Next Article in Journal
A Low Memory Requirement MobileNets Accelerator Based on FPGA for Auxiliary Medical Tasks
Next Article in Special Issue
Assessment of Remote Vital Sign Monitoring and Alarms in a Real-World Healthcare at Home Dataset
Previous Article in Journal
An Efficient Approach to Predict Eye Diseases from Symptoms Using Machine Learning and Ranker-Based Feature Selection Methods
Previous Article in Special Issue
Respiratory-Induced Amplitude Modulation of Forcecardiography Signals
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:

Advancement in the Cuffless and Noninvasive Measurement of Blood Pressure: A Review of the Literature and Open Challenges

Mohammad Mahbubur Rahman Khan Mamun
1,* and
Ahmed Sherif
Department of Electrical and Computer Engineering, Tennessee Technological University, Cookeville, TN 38505, USA
School of Computing Sciences and Computer Engineering, The University of Southern Mississippi, Hattiesburg, MS 39406, USA
Author to whom correspondence should be addressed.
Bioengineering 2023, 10(1), 27;
Submission received: 10 December 2022 / Revised: 20 December 2022 / Accepted: 21 December 2022 / Published: 24 December 2022
(This article belongs to the Special Issue Monitoring and Analysis of Human Biosignals)


Hypertension is a chronic condition that is one of the prominent reasons behind cardiovascular disease, brain stroke, and organ failure. Left unnoticed and untreated, the deterioration in a health condition could even result in mortality. If it can be detected early, with proper treatment, undesirable outcomes can be avoided. Until now, the gold standard is the invasive way of measuring blood pressure (BP) using a catheter. Additionally, the cuff-based and noninvasive methods are too cumbersome or inconvenient for frequent measurement of BP. With the advancement of sensor technology, signal processing techniques, and machine learning algorithms, researchers are trying to find the perfect relationships between biomedical signals and changes in BP. This paper is a literature review of the studies conducted on the cuffless noninvasive measurement of BP using biomedical signals. Relevant articles were selected using specific criteria, then traditional techniques for BP measurement were discussed along with a motivation for cuffless measurement use of biomedical signals and machine learning algorithms. The review focused on the progression of different noninvasive cuffless techniques rather than comparing performance among different studies. The literature survey concluded that the use of deep learning proved to be the most accurate among all the cuffless measurement techniques. On the other side, this accuracy has several disadvantages, such as lack of interpretability, computationally extensive, standard validation protocol, and lack of collaboration with health professionals. Additionally, the continuing work by researchers is progressing with a potential solution for these challenges. Finally, future research directions have been provided to encounter the challenges.

Graphical Abstract

1. Introduction

More than a billion people suffer from hypertension in the whole world [1]. Hypertension increases the risk for life, endangering cardiovascular issues and severe diseases such as stroke, renal and kidney failure, etc. If unnoticed and not managed, hypertension can lead to serious health complications and may eventually deteriorate to the patient’s death. If monitored, hypertension and hypotension can be managed using lifestyle modification, change in food habits and medication suggested by a physician [2,3]. The problem is, unlike many major diseases in humans, abnormally high or low blood pressure may go unnoticed due to a lack of monitoring and lack of significant symptoms. The best way to detect this issue is to regularly monitor using a reliable method and keep track of the changes [4,5]. In every clinical guideline, hypertension and hypotension management constitute significant sections; for example, systolic blood pressure over 130–140 mmHg is categorized as stage 1 hypertension [6]. Diastolic blood pressure (DBP) below 60 mmHg and systolic blood pressure below 90 mmHg is called hypotension, which means below normal blood pressure (BP). Hypotension inflicts oxygen deprivation on different parts of the human body; tissue death happens because of that. Hypotension symptoms include but are not limited to dizziness, general pain, sluggishness, etc., which may end in severe consequences such as cardiac arrest; on the other hand, hypertension symptoms span from fatigue and shortness of breath to heart attack, stroke, and organ failure [7,8]. Prehypertension starts with SBP between 121–139 mmHg and DBP between 81–89 mmHg [9]. Although this is not staged one hypertension, prehypertension is within a range with the potential risk of going over to stage 1. SBP between 140–159 mmHg is hypertension 1, 160–179 mmHg is hypertension II, and over 180 mmHg is called hypertensive crisis [10,11,12,13,14]. During hypertension, the stress over the heart increases; although medical treatment may not be mandatory at this stage, frequent monitoring is necessary for a sudden spike during the daytime. During hypertension II, without monitoring and timely treatment, the higher pressure may result in organ failure and irreversible damage to different body parts. The hypertensive crisis may result in numbness of body parts, vision problems, breathing difficulty, etc. Eventually, a heart attack or stroke may occur if left untreated.
The research revolving around monitoring BP mainly follows two approaches; one is to try to make the existing cuff-based measurement system more user-friendly or adjustable to day-to-day life; another is to use biomedical signals which are easily accessible to find a technique that will measure BP within reasonable accuracy. Since the first approach’s accuracy level is already reasonable, making it smaller, more efficient, and robust are the main objectives of newer devices. In the second approach, reliability or accuracy is the biggest concern and versatility in terms of age and other factors. So, that approach emphasizes those since acquiring the needed biomedical signals already helps to keep those designs smaller while removing the necessity of external stimulations such as pressure on the artery. The advancement of sensor technology, signal processing techniques, PCB design, and artificial intelligence heavily influences the second approach. Stephen Hales did his horse’s first blood pressure measurement using an invasive method [15,16]. Later, the process was improved by physicians for human use while keeping the technique invasive [17]. The first catheter was implanted in 1949 into the artery of humans; this technique allows us to measure the BP recording without further discomfort [18]. Later this catheter technique was further improved by another scientist so that a wide variety of blood pressure could be observed [19,20,21]. In terms of accuracy or access to the wide variation of change in BP, the invasive technique is superior to all other techniques, but unless the patient requires urgent attention or in case of a routine checkup, this method is not feasible due to risk for infection, insertion discomfort, specialist to perform the procedure, blood loss and structural damage, etc. [22,23,24]. Because of those reasons, the research for noninvasive techniques progressed further in the initial period based on the cuff-based method, widely used at home and health centers [25,26,27,28,29]. Although the cuff-based method became less cumbersome due to the invention of a user-friendly device, the discomfort or the necessity of the cuff is not removed. That introduced the possibility of using other means to measure blood pressure while keeping the accuracy and reliability the same as before with conventional methods. Currently, a massive amount of medical data is available for data analysis; at the same time, the advancement of sensor technology allows us to acquire vital signs with precision and ease. Additionally, physiological data such as electrocardiography (ECG) and photoplethysmography (PPG) can be used for biomedical signal analysis [30]. All this can be brought together to invent new techniques or methods to measure blood pressure in a noninvasive and cuff way with the application of artificial intelligence [31,32,33,34,35]. Although theoretically, it is possible on paper since the data acquisition, design, signal processing, machine learning, etc., all are readily available to implement. However, till now, none of the techniques have become a reality with a reliable, accurate, and robust device which can replace the conventional technique.

1.1. Current Literature Survey

Several review papers have been available related to cuffless blood pressure measurement. Some of them focus on features from the PPG signal [36,37]; on the other hand, some encapsulate human physical activities to correlate them with BP variation [38,39]. A review on cuffless blood pressure monitoring has been done in this article [40], focusing on the necessity of measuring BP more frequently and continuously. The review also discussed the new sensors and algorithms incorporating machine learning. Another survey was done to incorporate the progress of narrow to deep learning methods for BP measurement while comparing the performance of PPG and ECG [41]. They have also compared different feature extraction techniques using PPG signal and included heart rate estimation along with BP. Shenda Hong et al. [42] discussed the opportunities and challenges of deep learning while using ECG signals for diagnosis purposes; during their discussion, they also discussed the possibility of using ECG signals to detect blood pressure measurement in a short subsection. One of the better literature surveys was done by Ramakrishna Mukkamala et al. [43], which encompasses cuffless blood pressure as a whole, including but not limited to historical methods, current cuff-based methods, and the research involving PPG, ECG signals. They have divided the study into calibrated and non-calibrated methods where features from PPG and ECG were put together into the calibrated method and use of oscillometric, volume control, and ultrasound into the uncalibrated method. Although other than PPG and ECG signals, there are other biomedical and physiological signals available in which researchers tried to get any relationship with the change of BP, one such literature review [44], which looked at all kinds of physiological signals, was done by Manisha Sharma et al. Ramakrishna Mukkamala et al. have done a review [45] emphasizing the cuffless wearable device and smart devices to measure blood pressure in the current market; they have also discussed a large number of publications or research happening where the conclusions are unreliable due to inadequate methodologies. In this review article [46], the future use of artificial intelligence in managing hypertension has been discussed; they have tried to examine the explainability of artificial intelligence (AI) to help health professionals understand the black-box nature of models. Kazuomi Kairo introduced the discussion [47] about using a miniature wearable monitoring device for remote measurement of BP; his primary focus was out-of-office BP measurement expecting dramatically change in the quality of hypertension management. Niklas Pilz et al. published a review [48] article discussing pulse wave velocity-related techniques for BP measurement and techniques involving deep learning to shortlist the advantages and limitations of different methods. Apart from these reviews above, some other literature surveys have been done [10,49,50,51] focused on ambulatory blood pressure monitoring, diagnosing hypertension, noninvasive techniques of BP measurement, contactless BP monitoring, etc.

1.2. Survey Goal

All the literature above focused on one or more aspects of BP measurement, from traditional invasive methods to current research involving noninvasive-cuffless techniques. The existing surveys still have some limitations, such as: using a limited number of studies or techniques and failing to cover both ECG and PPG signals to extract features for BP measurement. However, with the ever-evolving nature of technology with signal processing techniques and artificial intelligence, there are scopes to contribute with a holistic literature review on BP measurement, adding significant scientific value. We have presented the literature survey in the following manner: first, a framework for the search and selectin process has been provided; second, conventional BP measurement techniques are summarized; the third, motivation behind the cuffless BP measurement techniques was discussed; fourth, we have also brought forth the technologies which are helping to get us from traditional method to cuffless BP measurement, fifth, the features which have been used for techniques involving artificial intelligence were discussed, sixth, all the machine learning and deep learning techniques which have been used for BP measurement in recent years are summarized, seventh, the available datasets were summarized, and finally the challenges and future way forwards were proposed so that AI can have more significant impact finding techniques for reliable and robust BP measurement. The two main questions we have tried to solve in this literature survey are: (1) How are the standard features from biomedical signals used in current studies to find noninvasive and cuffless BP measurements? (2) How have artificial intelligence techniques (machine learning and deep learning) been proposed in recent research works for the same purpose, and which research directions need attention? To answer these research questions and understand the premise, we have covered traditional BP measurement techniques, cuff-based techniques, and the eventual use of AI for cuffless BP measurement.

2. Framework for the Search and Selection Process

To incorporate a thorough review, the search procedure was implemented using machine learning and deep learning with features from biomedical signals to generate a model for measuring cuffless noninvasive BP. For this purpose, google scholar and PubMed were used with a timeline from 2010 to 2022. The following general search term was used from both database: (“ECG” or “electrocardiogram,” “ECG” or “EKG” or “PPG” or “photoplethysmography” or “biomedical signal,” “vital signs” or “physiological signals” or) and (“cuffless blood pressure” or “noninvasive blood pressure” or “hypertension” or “blood pressure” or “arterial pressure” or “hypotension” or “BP”) or (“machine learning” or “deep learning” or “artificial intelligence” or “neural network” or “convolutional neural network” or “ANN” or “CNN” or “long short term memory” or “LSTM” or “RNN” or “recurrent neural network”) or (“pulse transit time” or “pulse wave velocity”). It is essential to point out that using those keywords, many articles were found that were not related to this literature survey, and those were sorted out at the initial stage. Figure 1 depicts the selection process, divided into four stages (identification, screening, eligibility, and included). Overall, 745 articles were retrieved from the search using google scholar and PubMed. After removing the duplicate ones, the number of articles remaining was 643. Those articles were screened for title and abstract, and 529 were removed for not being relevant to this study or out of scope. The remaining 114 full-text articles were checked thoroughly, and 49 more were removed due to different reasons such as missing evaluation or comparison, insufficient dataset or sample number, lack of rationale behind decisions, insufficient explanation of experiment or method, etc. Finally, 65 articles were found relevant to this literature review and fulfilled the criteria in the eligibility section of the framework.

3. Traditional Techniques for BP Measurement

Among all the techniques, invasive and noninvasive arterial catheterization is the gold standard for accuracy. A thin and hollow tube is placed or inserted into the patient’s artery to perform this measurement technique. In the blood vessel, a manometer has been placed and used to measure the waveform of blood pressure change [52]. Blood pressure change can be tracked from the waveform and other cardiac activity [53]. This technique has been prevalent to frequently measure an ill patient’s BP reading (mainly in the incentive care unit). In case of hypotension, the patient can be given intravenous fluids to increase blood pressure immediately.
On the other hand, actions can be taken to lower blood pressure in the case of hypertension immediately. In both cases, frequent measurements are necessary to take urgent measures to avoid a more deteriorating conscience. The downside of these techniques is pain during the placement of the needle stick and catheter during insertion, potential infection in the place of catheters inserted, blood clots or bleeding during that time, etc. In Figure 2, traditional methods of noninvasive BP measurement techniques are shown. Among the noninvasive techniques for BP measurement, manual auscultation is an indirect technique where BP estimation is the most accurate [54]. This technique was initially invented in 1896 but later improved by Nikolai Korotkoff by placing a stethoscope over the brachial artery [55,56]. The blood flow sound disappears and reappears as the cuff is inflated and deflated. Using a sphygmomanometer, the measurement of SBP and DBP is done. The process requires an experienced clinician for reliable and consistent measurement [38]. Additionally, due to the toxicity of mercury in sphygmomanometers in many counties, the use of this tool is restricted. Additionally, due to the inflation of cuffs, considerable pressure is being put on the patient, which is not convenient for frequent measurement [57]. Among other BP measurement techniques, the oscillometric method is most common among manufacturers due to the emergence of digital electronics [58]. It measures mean arterial pressure and prepares estimates of SBP and DBP. The process starts with a cuff placed over the brachial artery in the upper arm; when the pressure increases through the cuff, the blood flow stops, and when the pressure releases, the pressure goes below MAP the pulsation decreases. From the oscillometric waveform in the time domain, the SBP and DBP were measured from MAP [38]. The downside of the oscillometric method is the uncertainty from the estimation of SBP and DBP, which does not reflect the standard expected from the Association for the Advancement of Medical Instrumentation (AAMI). There have been many variants of this technique using different signal analysis techniques, but still, the recommendation of AAMI is not met with consistency [59,60,61,62,63].
Using the modified Peaz principle, the vascular unloading technique is based on [64]. To maintain a constant blood volume, the pressure inside the artery is linearized by the outside pressure from the cuff placed on the finger [65]. The cuff consists of an infrared light source and photodiode to measure the blood volume in the arteries; the cuff’s pressure varies to maintain a constant blood volume. This intraarterial pressure is equal to finger cuff pressure determined by a manometer. Although continuous BP waveform can be generated with no chance of infection using this method, wearing a cuff for a long time is not convenient. Additionally, cuff size changes with finger size, and there is the possibility that finger pressure can be different compared to the considerable artery pressure making the measurement less reliable. Using arterial tonometry, a fair amount of pressure is applied on the finger over the radial artery to deform the shape of the finger [66]. The area where the measurement takes place should have a bone underneath the artery [67,68]; the pressure in the artery is proportional to the vertical displacement or bend in the finger [69]. The displacement is proportional to the arterial blood pressure waveform [70,71]. The shortcomings of this technique are a requirement of a specific type of place for measurement, and the position of that place must be very steady, making it unsuitable for more prolonged or frequent measurements. Additionally, placement accuracy is critical, etc.

4. Motivation for Cuffless Measurements of BP

This section discusses the motivation behind the cuffless blood pressure measurement. The discussion starts with the existing problem with traditional measurement techniques. After that, the current development of BP measurement devices will be summarized, and the rationale for research with the cuffless method must be examined. Additionally, the biomedical signals which are easily accessible and can be used for BP measurement in a cuffless manner will be discussed.

4.1. Limitations with Traditional Noninvasive Methods for BP Measurement

Although the measurement from the invasive method is direct from the blood flow in the artery, all the noninvasive measurement techniques are indirect. Thus accuracy and consistency are difficult to match. As per research, manual auscultation estimates the SBP lower than the invasive method and the DBP higher than the invasive method [72,73]. Many other techniques, such as oscillometric, auscultatory, and volume clamp devices, are calibrated using the invasive method. However, even with the calibration due to the chosen artery for measurement, the result still can be inaccurate [74,75,76]. Among all the conventional noninvasive methods, oscillometric devices dominate the clinical practice and the central portion of the database with BP measurement [77]. There are several limitations with the noninvasive methods based on the cuff. First, when it comes to the manual auscultation method, the cuff-based method requires trained professionals to perform the BP measurement accurately and reliably. Second, the cuff-based device is not readily available to everyone needing frequent measurements. Especially for the area where resources are limited, having an individual device for measurement is scarce, which is the main reason behind the awareness about own BP reading [78]. Third, during the measurement, the inflation and deflation of the cuff make inconvenience for the patient. Additionally, the patient must be still and relaxed during measurement, which is also tricky for frequent measurements [79,80]. Especially during nighttime, it is very cumbersome to have the cuff works frequently and hinders ambulatory measurement [81,82,83,84]. To solve these problems, the cuff must be removed from the setup while maintaining accuracy and reliability. The added functionalities needed are the ability to have frequent measurements with immediate results, a less cumbersome setup for measurement, the ability to provide user-friendly measurement techniques, and the inexpensive.

4.2. Necessity for Validation Protocol for Existing and Future BP Measurement Device

Among all the necessary clinical tests, measuring BP is one of the important ones, so both overestimation and underestimation severely impact the treatment process for the patient [85]. Usually, the most common error happens with overestimating BP measurement, which can cause issues such as wrong or untimely medications, unnecessary use of resources, and visits to physicians, etc. [86,87,88]. For example, hypertension prevalence impacts around thirty percent with a consistent inaccuracy of 2.7/5 mmHg [89]. Along with World Health Organization, several international organizations showed concern about validating the existing BP measurement device [90,91,92,93]. The main proposal was to strengthen the requirements for validation protocol and make it difficult for unvalidated products to be used or authorized to be sold. This proposal includes the device and new technology to measure BP, such as noninvasive cuffless methods. The World Health Organization (WHO) has presented the required specifications for automated noninvasive BP measurement devices with cuffs only; it encapsulates the ambulatory nature of measurement at home or outside and even in clinical setup [94]. The suggestions provided for BP measurement devices are, first, there has to be a solid regulatory board or office to ensure that only validated devices are available in the market [95]. Second, there must be a process for validating the device without bias. Third, manufacturers should disclose in detail the validation process they have performed before making the product available [95]. Fourth, the government should take necessary actions to make the validated product available to low resources areas and populations [54,96]. Fifth, to provide necessary training to professionals to differentiate between validated and not validated devices. Although the concerns from the WHO’s point of view are towards the cuff-based method, they have also targeted the research using the cuff-less method. Since some available devices work based on the cuffless method, those products should also be considered under the same validation protocol. Since all noninvasive methods are indirect estimations, validation should be critical while preparing or proposing any technique.

4.3. Improvement in Technology Helping Research with Cuffless BP Measurement

In this subsection, the emergence of the latest technologies enables the research of using biomedical signals to bring forth possible techniques to measure BP. Figure 3 depicts the technological advances responsible for the research progress in cuffless BP measurement techniques. The significant advancements are smaller sensors to acquire biomedical signals, improved signal processing techniques, highly efficient and robust calculation power, etc. [97]. First, the biomedical signals that are now being used in research will be discussed, then the adaptation of these sensors in convenient devices will be discussed. The ECG measures the heart’s electrical activity, which can be acquired by using electrodes placed in specific areas of the human body near the heart. A typical ECG wave consists of P, Q, R, S, and T peaks. The rhythmic activity acquired by ECG provides vast information about the structure and functions of the heart [98,99]. Different parts of the ECG wave can be correlated with the change of BP change after doing signal analysis techniques over the ECG wave [99,100]. Studies have shown the relation between changes in SBP and DBP with interval changes such as PR and ST segment [101,102].
PPG signal provides the amount of oxygen saturation in blood using the calculation from transmitted and receiving light emitted from led in a specific area on the body such as the finger or ear [103,104]. Since blood pressure in arteries or blood vessels impacts the oxygen saturation or peak information or interval information from consecutive PPG waves, this signal has been extensively used in various research to find BP measurements [37,105]. When the blood is ejected from the heart, the body goes through changes such as displacement, acceleration, and these measurements can be recorded as ballistocardiogram (BCG) [106,107,108]. Figure 4 shows standard biomedical signals related to BP measurement techniques. Among other biomedical signals using ultrasound measuring absolute blood volume, blood volume oscillations measured by electrical bioimpedance and seismocardiography (SCG) measuring the lower frequency vibrations from the heart are also used in many research studies to make a relation with the change in blood pressure [108,109,110]. Due to the advancement in smartphones concerning a higher capacity for calculation and image processing ability—the smartphone can be used to create a PPG wave, and various information can be generated from that wave as a result [111,112]. Additionally, recent high-end smartphones have additional sensors to acquire different biomedical signals, such as SCG, PPG, ECG, and BCG, etc., to provide health alerts based on threshold strategies [113,114]. Like smartphones, smartwatches and fitness trackers are equipped with advanced sensor technologies and analysis capabilities to provide alerts for different health conditions [115,116,117,118]. To summarize, the recent technological advancement helped researchers have easy access to biomedical signals and high calculation capabilities in terms of speed and complexity.

4.4. Existing Wearable Device for Hypertension Management

The necessity of measuring blood pressure more frequently comes from excessive BP levels in the morning has been proven to be related to the risk of stroke, brain hemorrhage, and other organ damage [120,121,122,123,124,125,126]. Additionally, the nighttime elevation of blood pressure level is consistent with different cardiovascular issues [127,128,129,130]. So, the traditional habit of measuring blood pressure once in a while must be changed. Especially for people with a heart condition and stressful daily life, frequent measurement is a must to detect any deteriorating scenario beforehand. Compared to traditional BP measurement devices are cumbersome, the latest wearable devices can provide better adaptability for the patient. However, nearly all of the devices currently available in the market which are wearable and capable of measuring BP (as claimed by the manufacturer) are unvalidated and unreliable in terms of accuracy [95,131]. These facts did not stop the manufacturer from bringing devices with the ability to calculate BP from biomedical signals to some extent. In this subsection, only the devices currently existing will be discussed. Two devices from Omron healthcare have been submitted for validation criteria to AAMI; they fulfill the validation criteria while the patient stays sitting [132,133,134,135]. Another device from Omron was submitted for validation which works with higher accuracy while the patient remains in a specific orientation [133]. Both submissions were based on the device following an oscillometric measurement technique on the wrist. Other than those, some more devices tried to validate but did not match the threshold set by AAMI [135,136]. Two devices were submitted for validation to IEEE standards but did not pass the AAMI standards [137,138]. Additionally, till now, although there has been a significant number of studies done using PPG to measure BP, none of those made any substantial contribution towards consistent and reliable measurements of arterial BP [139]. Some devices combined PPG and ECG to make the measurement more robust, but even those did not pass validation with standardization organizations [140,141]. Some other approaches used other biomedical signals, such as bioimpedance with ECG but failed to uphold the criteria of standardization organization as a generalized rule [142,143]. Although wearable BP technology can potentially improve symptom detection for cardiovascular diseases, more research must be done before those can be used in real scenarios.

4.5. Rationale behind the Necessity to Continue Research on Cuffless BP Measurement

Research on cuffless blood pressure and measurement has recently increased rapidly [144]. At the same time, different devices claimed to be able to measure blood pressure using different types of biomedical signals are becoming available [144,145,146]. So, with the emergence of a high number of research and devices in the market, validating the techniques and devices is becoming increasingly important [74,147,148]. There are several standardization protocols for BP measurement, such as AAMI British Hypertension Society Protocol (BHS), and none were intended for cuffless blood pressure measurement [95,138,140,149]. So, until there is a universally acceptable validation protocol specifically for cuffless BP measurement methods and devices, it is not easy to accept the result or evaluation of techniques from any research, irrespective of how accurate or reliable the method claims to be.
Another problem is using a different number of samples and different kinds of datasets for training or testing the models. The validation protocol needs to provide specific information about evaluating the method, which can be compared or replicated easily [150,151]. Another challenge for the cuffless BP measurement techniques, which are calibration-free, is the significant variation that occurs in BP during the whole day due to lifestyle or daily work habits. Finally, most recent studies work with a trained model that compares different features with the change in BP. However, the actual relationship between the features and BP change is still unclear, especially for the techniques or devices that are not calibrated [152,153,154,155,156,157,158,159].

5. Features from Biomedical Signals for BP Measurement

During the last decades, numerous research has been done to find a new method for cuffless BP measurement, the rationale behind the choice of topic was the irritation and inconvenience caused by cuff based method. Additionally, the existing cuff-based method is unsuitable to have frequent measurements of BP. Among all the biomedical signals the scientist chose to work with, the most research involved photoplethysmography (PPG) signal [160]. In this section, the popular features from the biomedical signals that were used for cuffless BP measurement techniques will be discussed. The objective of the following discussion is not to summarize the research or articles producing a result but rather to summarize whether the results are consistent enough to be accepted.

5.1. Pulse Transit Time and Pulse Arrival Time

Pulse transit time is defined by the time it takes for a pulse pressure wave to travel from one point of the artery to another, as shown in Figure 5 [36]. To measure the PTT, the most common method is to place two PPG sensors in two places close to the artery: ear, finger, toes, etc. [161,162,163]. Research has shown that there has been a reciprocal relationship between blood pressure and pulse transit time [164,165]. On the other hand, a researcher has also proved that only PTT is not sufficient to measure accurate BP levels [164,166].
Along with PPG, the ECG wave has also been used to measure PTT. In this case, measuring PTT means ECG is considered the proximal waveform, and another PPG sensor is used for the distal waveform to calculate PTT. Values of PTT from a specific patient and his/her BP can be related using a model based on empirical data. However, the problem is that the data needs to be calibrated frequently due to vascular aging. The correlation between PTT and BP levels is good but not strong enough to have a conclusive result; mostly, the correlation value remains between 0.7–0.8 [168,169,170,171]. Till now, there has been no proof of any PTT-based method which can be used as a general formula or method for BP measurement, the most common flaw is that once the model is evaluated using a dataset other than the one used for producing the technique, the evaluation fails to uphold any standard protocol (AAMI or BHS) [172,173].
Similar to PTT, there is another parameter that is popular among researchers is pulse arrival time. It is the time difference between heart activation and pulse wave at a distal point such as a finger. PAT can also be defined as pre-ejection delay plus the PTT [36]. Usually, the peak of the ECG wave is used as the proximal point, and another PPG sensor is a distal point to measure PAT [174,175,176]. Similar to PTT, the PAT is also influenced by different physical attributes of the patient and shows similar limitations using it in the model for BP measurement [172,173]. Studies proved PAT to be better-suitable than PTT but require calibration for individual use.

5.2. Pulse Wave Velocity

Pulse wave velocity (PWV) is the pulse wave’s velocity that travels from a proximal point to a distal point through a length L [29,177]. The relation between PTT and PWV is reciprocal. Because the elasticity of the artery changes from person to person, even with the age of the same person, the length of L will also change from person to person. Despite these issues, researchers tried to find the relationship between PWV and BP levels (SBP and DBP) [178,179].
PWV = L PTT = artery   wall   thickness × Elasticity   of   artery   wall blood   density × distance
The relationship can be illustrated using the Moens-Kortweg equation, as shown above [180]. The existing proposed techniques require extensive calibration, even for a single person. Since the model scientist produced can not be generalized, and the calibration requirement frequently uses a cuff-based method, no clinically usable device or end product is available based on PWV [181,182].

5.3. Pulse Wave Analysis

Pulse wave analysis means using the pulse wave through signal processing techniques and extracting necessary features for further use in building a model for BP measurement. A typical step for pulse wave analysis to get a BP measurement model is shown in Figure 6. The process requires one or more sensors to measure PPG waves, signal processing tools for removing noise and artifacts, feature extraction and selection, etc. Many attempts have been made to build a model for BP measurement using pulse wave analysis [159,183,184,185]. Examples of features in a typical PPG wave are shown in Figure 7.
Four main types of features are popular among researchers to extract from pulse waves for cuffless BP measurement: time domain-based, frequency domain-based, time-frequency domain based, and statistically based, as shown in Figure 8. Some researchers opted for single-domain features, while others opted for a combination of several domain features. A typical pulse wave analysis diagram is shown in Figure 6. Processing raw signal for noise or artifact are not in the scope of this literature review; rather, the focus will be the typical features used by the researchers from pulse wave analysis to create a model for BP measurement. As shown in Figure 7, a common strategy is to get the derivatives of the PPG signal to get different peak values and intervals. Additionally, using the Fourier transform, the primary information can be achieved at significant frequencies. Other than the physical features such as age, height, weight, BMI, heart rate, etc., another aspect researchers follow for features are statistical features [187,188,189]. The time domain, frequency domain, statistical domain, and physical features used in the literature survey have been summarized in Figure 8 [190,191,192,193,194,195].

6. Machine Learning and Deep Learning in Cuffless Noninvasive Blood Pressure Measurement

This section will discuss machine learning and deep learning in the studies and research to measure cuffless noninvasive blood pressure. The focus of the discussion will not be on how individual techniques proposed by the researcher performed but on how and which algorithms were implemented and their respective shortcomings. The rationale behind the choice of focus is, comparing or summarizing performance will not add any value since there is no general protocol available for validation of the technique for cuffless noninvasive BP measurement. So, the literature survey will be more effective and informative if it involves how the algorithms were used.

6.1. Conventional Machine Learning Algorithms

Conventional machine learning algorithms involve manual extraction and feature selection. Starting with the raw signal, preprocessing the signal, extraction of features, and selection of features, the finalized features are used to train the machine learning algorithms to create a model. The most popular machine learning algorithm used widely is the regression algorithm [189,196,197,198,199,200,201,202,203]. The regression algorithm can be in several forms: linear regression, least square regression, lasso or ridge regression, etc. To summarize how the regression algorithms were implemented, most studies used a PPG signal or a combination of PPG and ECG signals to extract the required features from those signals. After getting features, the usual process was to take the most valuable and relevant features and use only those to train the regression model. Later the regression models were evaluated using the same dataset. Nearly all the studies followed this same basic format; the changes from one study to another were mostly in the choice of signal, several features to be extracted, the algorithm used to optimize features, and the sample number for training and testing. The typical limitations found were the inability to reproduce performance with a separate database other than the one used for training a small number of samples to train with. The way forward for regression-based analysis is to use a large dataset, and more research needs to be done before claiming to achieve any significant and usable outcome. Similar to regression algorithms, the research used a support vector machine (SVM), AdaBoost, random forest, and K-nearest neighbor, etc., to find a model to measure BP [185,204,205,206,207,208,209,210,211]. Additionally, the limitations were almost the same for these conventional machine learning algorithms. Although some studies claimed that one of the conventional machine learning algorithms performed better than others, their performance was not up to the mark or consistent regarding following standards such as AAMI or BHS. Based on the literature survey, it is impossible to agree with any claim about the superiority of any conventional machine learning algorithm over others.
Although numerous publications have used conventional or shallow machine learning architecture to find models to measure BP, none of these came into reality or were of clinical use. If researchers want to pursue this path for a solution, they need to make sure to do the following to make their solution sustainable and reliable:
  • The model must reproduce the claimed result using other datasets or real-time data.
  • The rationale behind the performance should accompany the use of algorithms.
  • The number of features and optimization algorithms needs to be explainable.
  • AAMI and BHS standards need to be passed for both SBP and DBP.
  • Since all the features are from biomedical signals due to physiological activities, calibration is mandatory. The purpose, frequency, and procedure of calibration need to be addressed.
  • The database (either online or real patient data) needs to be well distributed regarding demographic data.
  • There has to be a generalized validation protocol in place so that the performance of all the techniques can be compared without any bias.

6.2. Deep Learning Algorithms

With the advancement of deep learning networks through artificial neural networks (ANN), convolutional neural networks (CNN), recurrent neural networks (RNN), Long and short-term memory (LSTM), etc., they are using different biomedical signals such as ECG and PPG into learning network to achieve a model for BP measurement becoming very popular among scientists. The main reasons are: compared to shallow machine learning algorithms, deep learning algorithms can extract features on their own and can do a better job in most cases by finding features that might have been missed in the case of shallow ML, also although it is computationally extensive the performance achieved from deep learning networks were much better than conventional ML algorithms. For the past few years, scientists have tried every possible deep-learning algorithm for BP waveform prediction, where the model automatically learns critical features. The ECG signal can be acquired using an electrode over a specific part of the upper torso; the PPG signal can be acquired from either ear or fingertip. The gold standard arterial blood pressure can be chosen from the cuff-based method or a radial artery catheter. Figure 9 depicts an example of a deep learning algorithm from start to finish.
Although preprocessing is a common step for any machine learning algorithm, there is always room for improvement during preprocessing steps, which can make an evident difference in the outcome of the model’s performance. An example of a flow diagram of a deep learning network is shown in Figure 9. Since the deep learning algorithm takes care of the feature extraction and assigning appropriate weight for the impact of each feature, the preprocessing step needs to ensure the raw data is free from noise, segmented for training; inputs are normalized, etc. Due to the temporal similarities between PPG and ABP, most studies have used PPG as an input signal for their respective deep-learning algorithms. With the invention of new deep learning algorithms, researchers have tried almost all the available methods, such as wavelet neural network, long short-term memory, ANN, U-Net, MultiResUnet, V-Net, GAN, etc., for BP measurement model [24,32,35,212,213], even some studies involved using a hybrid model where multiple algorithms were used together for better performance [10,35,214,215,216,217,218,219,220,221,222,223,224]. A study was conducted to measure blood pressure by combining information from waveform (ECG and PPG), patient physical information, and features such as PTT and passing then as input of a neural network [225]. They have found that using features combined with deep learning algorithms worked better than using neural networks alone. Since all these studies were data-driven and the deep learning model follows a black box-themed approach, the only way the advantage or disadvantages can be discussed is based on input data processing and output data performance. The most common disadvantage of using deep learning networks was extensive computational complexity, lack of rationale behind how and why the architecture worked, a small number of subjects, use of the same dataset for training and testing, etc. On the plus side, nearly all those algorithms performed better than the conventional shallow machine learning algorithms. Compared to the scenario where shallow machine learning algorithms failed to pass the AAMI and BHS threshold, there are cases in recent years where the use of deep learning algorithms resulted in a ‘pass’ in AAMI or grade “A” or “B” in BHS standards [24,33,34,213,226].
In this section, the summarization or discussion of the individual study was intentionally avoided due to the following facts:
  • Although deep learning models provided better accuracy compared to shallow machine learning algorithms, the number of studies that matched the AAMI and BHS standards is deficient.
  • Since the approaches are mainly data-driven, the performance using the author’s choice of the dataset for their model should match using another random dataset. Nearly all the studies used the same dataset for test and validation.
  • Since deep learning algorithms require large samples to be adequately trained, there is a high chance that all the experiments using microscopic subjects may result in overfitting the technique.
  • There is no general validation approach available to compare performance among different studies.
  • No clinically acceptable method was produced or implemented to have a gold standard using deep learning techniques.
Although deep learning network does not require the manual intervention of feature extraction or feature optimization, these models are heavily computationally extensive. On the other hand, due to the lack of a global standard dataset, experiment protocol, or validation protocol, this research area is now limited to plugging biomedical signals, especially PPG and ECG, into different types of deep learning algorithms and their hybrid versions.

7. Dataset Used for BP Measurement Model

Among all the databases used in studies to make machine learning models for cuffless noninvasive BP measurement using biomedical signals, the most common ones are the University of Queensland’s vital signs dataset [227] and MIMIC [228]. Since deep learning requires a much larger dataset, most of the dataset is used for shallow machine learning algorithms, and nearly all private datasets can not be used for deep learning algorithms due to being too small. The following table contains the most popular database, the number of data, the signal from the database, and a short description. Since there is no general guideline or protocol available to create a database for the BP measurement model, the database has some disadvantages coming along with it. Table 1 shows different datasets being used in BP measurement models.

8. Challenges and Future Recommendations

The continuing research to get cuffless and noninvasive BP measurement model are facing challenges such as data collection, the accuracy of control data, standardization of public dataset, need for validation protocol, patient-specific issues, calibration, the efficiency of the algorithm, integration with the traditional method, issues with the specific dataset, deployment issues, and collaborations, etc. Along with discussing the challenges, there will be recommendations and ways forwards for the researcher to pursue their investigation.

8.1. Data Collection and Accuracy of Control Data

Until now, there is no standardized protocol for data collection to use the studies to make a BP measurement model. Different experiments have used different protocols while collecting data; on top of that, the dataset extensively used in research also employed different protocols. Additionally, if the objective is to prepare a model to be used by general populations of all ages, the training data needs to be collected, keeping that in mind. Most of the online dataset was collected during the patient’s intensive care unit stay, which can be impacted by many variations such as drug, type of disease, etc. Additionally, data acquisition should be made in a stationary position since any movement may add noise and unnecessarily change the rhythm. The protocol should be implemented so that, initially, the learning should start using static data, and gradually the learning should adapt to more dynamic situations.

8.2. Validation Protocol

There have been a significant number of studies done on this topic. However, due to a lack of similarity between the dataset, chosen data sample, or data acquisition protocol, it is tough to compare one technique. Some of the techniques are mathematical equation based, and some are machine learning algorithm based; the result from both may claim higher accuracy, but there is no way to compare which algorithm performed better. While working with machine learning algorithms, popular datasets contain ABP waveform to consider as the gold standard. When the dataset is privately owned or created from real-life patients, the number of subjects usually remains very small, and at the same time, getting ABP is difficult. To solve this problem, there must be a documented protocol for acquiring data from the patient, selecting the patient, preprocessing the data, and testing the data after preparing the model. Till now, there has been no such protocol available. Additionally, even in the successful experiments where a specific algorithm performed well, none of these have gone through a clinical trial. So the research speed needs to be matched with making the innovation available to be used clinically.

8.3. Calibration

One of the most difficult challenges for the current research work in this area is to produce a standard calibration approach that will not only provide a person-specific reliable BP model but also get updated to continue providing results with similar accuracy in the future. Researchers have tried to use a large dataset with sufficient variation and a wide range of data to train the model. However, no attempt replaced the necessity to calibrate using a standard method from time to time. There have been several attempts to calibrate the model, or in other words, fine-tun the parameters; such an attempt was with PTT calibration using data every 24 h using photoplethysmography intensity ratio [234], some other calibration attempts involved sphygmomanometer data [235,236]. However, none of the calibration methods is practical to implement or remove the need to use standard gold measurement systematically. The problems are twofold. First, a calibration need is due to a person’s physiological change. Without calibration, different health parameters’ impact on blood pressure change will go unnoticed. Second, a model prepared for a specific group of patients’ data can not be used as a general protocol unless it contains a calibration segment that adjusts the model parameter for another group of persons. So, preparing a model is only a part of the solution; when it comes to making the model usable clinically for the mass population, there has to be a standardized solution, including calibration steps that will not require the patient to use the cuff-based method frequently.

8.4. The Model or Technique Needs to Be Interpretable

Artificial intelligence in machine learning and deep learning dominates the current research endeavors regarding noninvasive cuffless BP measurement. With the advance of deep learning algorithms, more and more studies are getting published with more advanced deep learning models and better accuracy. However, there is a considerable gap to fill when explaining the necessary knowledge to elaborate on the model and its relation to the result. Without a proper explanation of the black box’ nature of the deep learning model to the audience, such as clinicians, physicians, health care workers, etc., there will always be an issue of trust. The researchers need to understand the necessity of creating a model that can be explained with a rationale to the people who will use the model confidently. Interpretability is a significant part of diagnosis and decision-making in the medical arena. Healthcare professionals must explain or at least understand how and why the model produces any specific result. Since feature selection approaches can be easily explained and related to medical professionals’ biological knowledge, researchers need to look for either keeping the deep learning model simple or using a hybrid model.

8.5. Model Deployment

The objective is to let the patient measure frequent blood pressure. However, the day is convenient, so using the small and portable device is mandatory to ensure the measurement setup is not cumbersome and hinders daily activities. Now, the majority of the research available only concerns themselves with only preparing a model and the performance of the model. However, very few tried or designed a small system that can replicate the model in real scenarios or work for actual people. The main challenges of deploying the model into real-world applications are. First, the signal acquisition in research mostly happens in the patient’s controlled environment and posture, which will not be the case in real life; second, the small and portable devices need the necessary calculation power to deliver accuracy and calibration capacity frequently. All the devices that can use the biomedical signal to measure BP are not certified or acceptable in clinical environments, so once researchers advance to deploying the models into the device, the challenges will come in a different form. Researchers may consider the challenges of possible deployment ahead of designing the model.

8.6. Necessity of Collaboration with a Health Professional

The research field needs collaboration between machine learning, signal processing expert, and health care professionals. To tackle the three main fronts of the solutions: acquiring proper signals, preparing a model, and deploying the model into the device, the collaboration can make a better plan. Right now, the majority of the studies are done involving machine learning experts only. The problem arises when the model’s interpretability and deployment come into question. The signals can be mixed with different kinds of noise, artifacts, and motion in an accurate word scenario compared to a lab scenario. Additionally, when it comes to deployment, the end device needs to be adaptable to the model and the calculation capacity needed by the model. Additionally, irrespective of the performance of the device or model, the professional medical needs to be confident about the result since they make the decision that can either improve or deteriorate patients’ health condition.

9. Conclusions

This work reviewed current development in cuffless and noninvasive blood pressure measurement techniques using biomedical signals. More specifically, the features from signals, machine learning, and deep learning were the main focus of this literature review. Although a significant amount of research happened with advanced signal processing and machine learning techniques, there are still scopes of improvement in consistency, calibration, interpretability, collaboration with medical professionals, deployment into a user-friendly device, etc. Since there is no device available now certified by an accepted regulatory body to be used clinically, the individual performance of the research was not the main focus of the survey. Instead, the focus was on the reasons behind the shortcomings and challenges. Unless the performance of the models developed by researchers using either an online dataset or a private dataset created manually can be replicated in an actual application and on a new patient, the goal of the research is to let people measure BP more frequently without cuff can not be materialized. Several recommendations and future research directions have been provided, such as: building an acceptable universal dataset for training models, making calibration without using cuff based method, creating a real-life scenario to account for all kinds of noise while acquiring signals, planning deployment feasibility in the research plan, making the models interpretable to physicians to gain confidence, etc. Unless these research paths are ventured extensively, the cuffless noninvasive BP measurement model using biomedical signal analysis should not be used in a clinical setup.

Author Contributions

Conceptualization, M.M.R.K.M. and A.S.; methodology, M.M.R.K.M.; formal analysis, M.M.R.K.M.; investigation, M.M.R.K.M.; writing—original draft preparation, M.M.R.K.M.; writing—review and editing, M.M.R.K.M. and A.S.; supervision, A.S. All authors have read and agreed to the published version of the manuscript.


This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Conflicts of Interest

The authors declare no conflict of interest.


  1. Ringrose, J.; Padwal, R. Wearable Technology to Detect Stress-Induced Blood Pressure Changes: The Next Chapter in Ambulatory Blood Pressure Monitoring? Am. J. Hypertens. 2021, 34, 330–331. [Google Scholar] [CrossRef] [PubMed]
  2. Bundy, J.; Li, C.; Stuchlik, P.; Bu, X.; Kelly, T.; Mills, K.; He, H.; Chen, J.; Whelton, P.; He, J. Systolic blood pressure reduction and risk of cardiovascular disease and mortality: A systematic review and network meta-analysis. JAMA Cardiol. 2017, 2, 775–781. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  3. World Health Organization. A Global Brief on Hypertension: Silent Killer, Global Public Health Crisis: World Health Day 2013; World Health Organization: Geneva, Switzerland, 2013. [Google Scholar]
  4. Siu, A.; US Preventive Services Task Force*. Screening for high blood pressure in adults: US Preventive Services Task Force recommendation statement. Ann. Intern. Med. 2015, 163, 778–786. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  5. Muntner, P.; Shimbo, D.; Carey, R.; Charleston, J.; Gaillard, T.; Misra, S.; Myers, M.; Ogedegbe, G.; Schwartz, J.; Townsend, R. Measurement of blood pressure in humans: A scientific statement from the American Heart Association. Hypertension 2019, 73, e35–e66. [Google Scholar] [CrossRef] [PubMed]
  6. Flack, J.; Adekola, B. Blood pressure and the new ACC/AHA hypertension guidelines. Trends Cardiovasc. Med. 2020, 30, 160–164. [Google Scholar] [CrossRef] [PubMed]
  7. Verhaeverbeke, I.; Mets, T. Drug-induced orthostatic hypotension in the elderly. Drug Saf. 1997, 17, 105–118. [Google Scholar] [CrossRef]
  8. Mamun; Khan, M.M.R.; Alouani, A. Myocardial Infarction Detection Using Multi Biomedical Sensors. In Proceedings of the 10th International Conference on Bioinformatics and Computational Biology (BICOB-2018), Las Vegas, NV, USA, 19–21 March 2018. [Google Scholar]
  9. Tewelde, S.; Liu, S.; Winters, M. Cardiogenic Shock. Cardiol. Clin. 2018, 36, 53–61. [Google Scholar] [CrossRef]
  10. Man, P.; Cheung, K.; Sangsiri, N.; Shek, W.; Wong, K.; Chin, J.; Chan, T.; So, R. Blood Pressure Measurement: From Cuff-Based to Contactless Monitoring. Healthcare 2022, 10, 2113. [Google Scholar] [CrossRef]
  11. Salkic, S.; Batic-Mujanovic, O.; Ljuca, F.; Brkic, S. Clinical presentation of hypertensive crises in emergency medical services. Mater. Sociomed 2014, 26, 12–16. [Google Scholar] [CrossRef] [Green Version]
  12. Giles, T.; Materson, B. Treating stage 2 hypertension. J. Clin. Hypertens. 2005, 7, 464–470. [Google Scholar] [CrossRef]
  13. Oparil, S.; Zaman, M.; Calhoun, D. Pathogenesis of hypertension. Ann. Intern. Med. 2003, 139, 761–776. [Google Scholar] [CrossRef] [PubMed]
  14. Iqbal, A.; Jamal, S. Essential Hypertension. In StatPearls; StatPearls Publishing LLC: Treasure Island, FL, USA, 2022. [Google Scholar]
  15. Tsyrlin, V.; Pliss, M.; Kuzmenko, N. The history of blood pressure measurement: From Hales to our days. Arter. Gipertenz 2016, 22, 144–152. [Google Scholar] [CrossRef]
  16. Lewis, O. Stephen Hales and the measurement of blood pressure. J. Hum. Hypertens 1994, 8, 865–871. [Google Scholar] [PubMed]
  17. Quinney, D. Daniel Bernoulli and the Making of the Fluid Equation; Keele University: Keele, UK, 1997; Available online: (accessed on 20 December 2022).
  18. Peterson, L.; Dripps, R.; Risman, G. A method for recording the arterial pressure pulse and blood pressure in man. Am. Heart J. 1949, 37, 771–782. [Google Scholar] [CrossRef] [PubMed]
  19. Athaya, T.; Choi, S. A Review of Noninvasive Methodologies to Estimate the Blood Pressure Waveform. Sensors 2022, 22, 3953. [Google Scholar] [CrossRef] [PubMed]
  20. Seldinger, S.I. Catheter replacement of the needle in percutaneous arteriography: A new technique. Acta Radiol. 2008, 49, 47–52. [Google Scholar] [CrossRef]
  21. E.C.PEIRCE. Percutaneous femoral artery catheterization in man with special reference to aortography. Surg. Gynecol. Obstet. 1951, 93, 56–74. [Google Scholar]
  22. McGhee, B.; Bridges, E. Monitoring arterial blood pressure: What you may not know. Crit. Care Nurse 2002, 22, 60–79. [Google Scholar] [CrossRef]
  23. Lam, S.; Liu, H.; Jian, Z.; Settels, J.; Bohringer, C. Intraoperative Invasive Blood Pressure Monitoring and the Potential Pitfalls of Invasively Measured Systolic Blood Pressure. Cureus 2021, 13, e17610. [Google Scholar] [CrossRef]
  24. Hill, B.; Rakocz, N.; Rudas, Á.; Chiang, J.; Wang, S.; Hofer, I.; Cannesson, M.; Halperin, E. Imputation of the continuous arterial line blood pressure waveform from non-invasive measurements using deep learning. Sci. Rep. 2021, 11, 1–12. [Google Scholar]
  25. Haque, C.; Kwon, T.; Kim, K. Cuffless Blood Pressure Estimation Based on Monte Carlo Simulation Using Photoplethysmography Signals. Sensors 2022, 22, 1175. [Google Scholar] [CrossRef] [PubMed]
  26. Xu, Z.; Liu, J.; Chen, X.; Wang, Y.; Zhao, Z. Continuous blood pressure estimation based on multiple parameters from eletrocardiogram and photoplethysmogram by Back-propagation neural network. Comput. Ind. 2017, 89, 50–59. [Google Scholar] [CrossRef]
  27. Miao, F.; Wen, B.; Hu, Z.; Fortino, G.; Wang, X.; Liu, Z.; Tang, M.; Li, Y. Continuous blood pressure measurement from one-channel electrocardiogram signal using deep-learning techniques. Artif. Intell. Med. 2020, 108, 101919. [Google Scholar] [CrossRef] [PubMed]
  28. Zakrzewski, A.; Huang, A.; Zubajlo, R.; Anthony, B. Real-time blood pressure estimation from force-measured ultrasound. IEEE Trans. Biomed. Eng. 2018, 65, 2405–2416. [Google Scholar] [CrossRef]
  29. Zakrzewski, A.; Anthony, B. Noninvasive blood pressure estimation using ultrasound and simple finite element models. IEEE Trans. Biomed. Eng. 2017, 65, 2011–2022. [Google Scholar] [CrossRef]
  30. Mamun, K.; Rahman, M.M.; Alouani, A. Automatic Detection of Heart Diseases Using Biomedical Signals: A Literature Review of Current Status and Limitations. In Proceedings of the Future of Information and Communication Conference, San Francisco, CA, USA, 3–4 March 2022. [Google Scholar]
  31. Landry, C.; Peterson, S.; Arami, A. Estimation of the Blood Pressure Waveform Using Electrocardiography. In Proceedings of the 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Berlin, Germany, 23–27 July 2019. [Google Scholar]
  32. Sadrawi, M.; Lin, Y.; Lin, C.; Mathunjwa, B.; Fan, S.; Abbod, M.; Shieh, J. Genetic deep convolutional autoencoder applied for generative continuous arterial blood pressure via photoplethysmography. Sensors 2020, 20, 3829. [Google Scholar] [CrossRef]
  33. Qin, K.; Huang, W.; Zhang, T. Deep generative model with domain adversarial training for predicting arterial blood pressure waveform from photoplethysmogram signal. Biomed. Signal Process. Control. 2021, 70, 102972. [Google Scholar] [CrossRef]
  34. Athaya, T.; Choi, S. An estimation method of continuous non-invasive arterial blood pressure waveform using photoplethysmography: A U-Net architecture-based approach. Sensors 2021, 21, 1867. [Google Scholar] [CrossRef]
  35. Ibtehaz, N.; Rahman, M. Ppg2abp: Translating photoplethysmogram (ppg) signals to arterial blood pressure (abp) waveforms using fully convolutional neural networks. arXiv 2020, arXiv:2005.01669. [Google Scholar]
  36. Mukkamala, R.; Hahn, J.; Inan, O.; Mestha, L.; Kim, C.; Töreyin, H.; Kyal, S. Toward ubiquitous blood pressure monitoring via pulse transit time: Theory and practice. IEEE Trans. Biomed. Eng. 2015, 62, 1879–1901. [Google Scholar] [CrossRef] [Green Version]
  37. Elgendi, M.; Fletcher, R.; Liang, Y.; Howard, N.; Lovell, N.; Abbott, D.; Lim, K.; Ward, R. The use of photoplethysmography for assessing hypertension. NPJ Digit. Med. 2019, 2, 1–11. [Google Scholar] [CrossRef] [Green Version]
  38. Ogedegbe, G.; Pickering, T. Principles and techniques of blood pressure measurement. Cardiol. Clin. 2010, 28, 571–586. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  39. Ding, X.; Zhao, N.; Yang, G.; Pettigrew, R.; Lo, B.; Miao, F.; Li, Y.; Liu, J.; Zhang, Y. Continuous blood pressure measurement from invasive to unobtrusive: Celebration of 200th birth anniversary of Carl Ludwig. IEEE J. Biomed. Health Inform. 2016, 20, 1455–1465. [Google Scholar] [CrossRef] [PubMed]
  40. Pandit, J.; Lores, E.; Batlle, D. Cuffless Blood Pressure Monitoring: Promises and Challenges. Clin. J. Am. Soc. Nephrol. 2020, 15, 1531–1538. [Google Scholar] [CrossRef] [PubMed]
  41. Maqsood, S.; Xu, S.; Tran, S.; Garg, S.; Springer, M.; Karunanithi, M.; Mohawesh, R. A survey: From shallow to deep machine learning approaches for blood pressure estimation using biosensors. Expert Syst. Appl. 2022, 197, 116788. [Google Scholar] [CrossRef]
  42. Hong, S.; Zhou, Y.; Shang, J.; Xiao, C.; Sun, J. Opportunities and challenges of deep learning methods for electrocardiogram data: A systematic review. Comput. Biol. Med. 2020, 122, 103801. [Google Scholar] [CrossRef] [PubMed]
  43. Mukkamala, R.; Stergiou, G.; Avolio, A. Cuffless blood pressure measurement. Annu. Rev. Biomed. Eng. 2022, 24, 203–230. [Google Scholar] [CrossRef]
  44. Sharma, M.; Rajput, J.; Tan, R.; Acharya, U. Automated Detection of Hypertension Using Physiological Signals: A Review. Int. J. Environ. Res. Public Health 2021, 18, 5838. [Google Scholar] [CrossRef]
  45. Mukkamala, R.; Yavarimanesh, M.; Natarajan, K.; Hahn, J.; Kyriakoulis, K.; Avolio, A.; Stergiou, G. Evaluation of the accuracy of cuffless blood pressure measurement devices: Challenges and proposals. Hypertension 2021, 78, 1161–1167. [Google Scholar] [CrossRef]
  46. Koshimizu, H.; Kojima, R.; Okuno, Y. Future possibilities for artificial intelligence in the practical management of hypertension. Hypertens. Res. 2020, 43, 1327–1337. [Google Scholar] [CrossRef]
  47. Kario, K. Management of hypertension in the digital era: Small wearable monitoring devices for remote blood pressure monitoring. Hypertension 2020, 76, 640–650. [Google Scholar] [CrossRef]
  48. Pilz, N.; Patzak, A.; Bothe, T. Continuous cuffless and non-invasive measurement of arterial blood pressure—Concepts and future perspectives. Blood Press. 2022, 31, 254–269. [Google Scholar] [CrossRef] [PubMed]
  49. Stergiou, G.; Mukkamala, R.; Avolio, A.; Kyriakoulis, K.; Mieke, S.; Murray, A.; Parati, G.; Schutte, A.; Sharman, J.; Asmar, R.; et al. Cuffless blood pressure measuring devices: Review and statement by the European Society of Hypertension Working Group on Blood Pressure Monitoring and Cardiovascular Variability. J. Hypertens 2022, 40, 1449–1460. [Google Scholar] [CrossRef] [PubMed]
  50. Huang, Q.; Yang, W.; Asayama, K.; Zhang, Z.; Thijs, L.; Li, Y.; O’Brien, E.; Staessen, J. Ambulatory blood pressure monitoring to diagnose and manage hypertension. Hypertension 2021, 77, 254–264. [Google Scholar] [CrossRef]
  51. Panula, T.; Sirkia, J.; Wong, D.; Kaisti, M. Advances in non-invasive blood pressure measurement techniques. IEEE Rev. Biomed. Eng. 2022. [Google Scholar] [CrossRef]
  52. Marino, P. The ICU Book; Lippincott williams & wilkins: Philadelphia, PA, USA, 2007. [Google Scholar]
  53. Wesseling, K.; Jansen, J.; Settels, J.; Schreuder, J. Computation of aortic flow from pressure in humans using a nonlinear, three-element model. J. Appl. Physiol. 1985, 74, 2566–2573. [Google Scholar] [CrossRef] [PubMed]
  54. Stergiou, G.; Palatini, P.; Asmar, R.; Ioannidis, J.; Kollias, A.; Lacy, P.; McManus, R.; Myers, M.; Parati, G.; Shennan, A. Recommendations and Practical Guidance for performing and reporting validation studies according to the Universal Standard for the validation of blood pressure measuring devices by the Association for the Advancement of Medical Instrumentation/European Society of Hypertension/International Organization for Standardization (AAMI/ESH/ISO). J. Hypertens. 2019, 37, 459–466. [Google Scholar]
  55. Peter, L.; Noury, N.; Cerny, M. A review of methods for non-invasive and continuous blood pressure monitoring: Pulse transit time method is promising? Irbm 2014, 35, 271–282. [Google Scholar] [CrossRef]
  56. Booth, J. A Short History of Blood Pressure Measurement. Proc. R. Soc. Med. 1977, 70, 793–799. [Google Scholar] [CrossRef] [Green Version]
  57. Shimek, J.; Emmanuel, J.; Orris, P.; Chartier, Y.; World Health Organization. Replacement of Mercury Thermometers and Sphygmomanometers in Healthcare: Technical Guidance; World Health Organization: Geneva, Switzerland, 2011. [Google Scholar]
  58. Geddes, L. Handbook of Blood Pressure Measurement; Humana Press: Totowa, NJ, USA, 2013. [Google Scholar]
  59. Liu, J.; Cheng, H.; Chen, C.; Sung, S.; Moslehpour, M.; Hahn, J.; Mukkamala, R. Patient-specific oscillometric blood pressure measurement. IEEE Trans. Biomed. Eng. 2015, 63, 1220–1228. [Google Scholar] [CrossRef] [Green Version]
  60. Forouzanfar, M.; Dajani, H.; Groza, V.; Bolic, M.; Rajan, S.; Batkin, I. Oscillometric blood pressure estimation: Past, present, and future. IEEE Rev. Biomed. Eng. 2015, 8, 44–63. [Google Scholar] [CrossRef]
  61. Babbs, C. Oscillometric measurement of systolic and diastolic blood pressures validated in a physiologic mathematical model. Biomed. Eng. Online 2012, 11, 1–22. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  62. Chandrasekhar, A.; Yavarimanesh, M.; Hahn, J.; Sung, S.; Chen, C.; Cheng, H.; Mukkamala, R. Formulas to explain popular oscillometric blood pressure estimation algorithms. Front. Physiol. 2019, 10, 1415. [Google Scholar] [CrossRef] [Green Version]
  63. Liu, J.; Hahn, J.; Mukkamala, R. Error mechanisms of the oscillometric fixed-ratio blood pressure measurement method. Ann. Biomed. Eng. 2013, 41, 587–597. [Google Scholar] [CrossRef]
  64. Wesseling, K.; Settels, J.; Wit, B. The Measurement of Continuous Finger Arterial Pressure Noninvasively in Stationary Subjects. In Biological and Psychological Factors in Cardiovascular Disease; Springer: Berlin/Heidelberg, Germany, 1986; pp. 355–375. [Google Scholar]
  65. Fortin, J.; Marte, W.; Grüllenberger, R.; Hacker, A.; Habenbacher, W.; Heller, A.; Wagner, C.; Wach, P.; Skrabal, F. Continuous non-invasive blood pressure monitoring using concentrically interlocking control loops. Comput. Biol. Med. 2006, 36, 941–957. [Google Scholar] [CrossRef] [PubMed]
  66. Butt, M.; Kazanskiy, N.; Khonina, S. Revolution in Flexible Wearable Electronics for Temperature and Pressure Monitoring—A Review. Electronics 2022, 11, 716. [Google Scholar] [CrossRef]
  67. Nelson, M.; Stepanek, J.; Cevette, M.; Covalciuc, M.; Hurst, R.; Tajik, A. Noninvasive measurement of central vascular pressures with arterial tonometry: Clinical revival of the pulse pressure waveform? Mayo Clin. Proc. 2010, 85, 460–472. [Google Scholar] [CrossRef] [Green Version]
  68. Chen, C.; Nevo, E.; Fetics, B.; Pak, P.; Yin, F.; Maughan, W.; Kass, D. Estimation of central aortic pressure waveform by mathematical transformation of radial tonometry pressure: Validation of generalized transfer function. Circulation 1997, 95, 1827–1836. [Google Scholar] [CrossRef]
  69. Matthys, K.; Verdonck, P. Development and modelling of arterial applanation tonometry: A review. Technol. Health Care 2002, 10, 65–76. [Google Scholar] [CrossRef]
  70. Salvi, P.; Grillo, A.; Parati, G. Noninvasive estimation of central blood pressure and analysis of pulse waves by applanation tonometry. Hypertens. Res. 2015, 38, 646–648. [Google Scholar] [CrossRef]
  71. Drzewiecki, G.; Melbin, J.; Noordergraaf, A. Arterial tonometry: Review and analysis. J. Biomech. 1983, 16, 141–152. [Google Scholar] [CrossRef]
  72. Picone, D.; Schultz, M.; Otahal, P.; Aakhus, S.; Al-Jumaily, A.; Black, J.; Bos, W.; Chambers, J.; Chen, C.; Cheng, H. Accuracy of cuff-measured blood pressure: Systematic reviews and meta-analyses. J. Am. Coll. Cardiol. 2017, 70, 572–586. [Google Scholar] [CrossRef] [PubMed]
  73. Prospective Studies Collaboration. Age-specific relevance of usual blood pressure to vascular mortality: A meta-analysis of individual data for one million adults in 61 prospective studies. Lancet 2002, 360, 1903–1913. [Google Scholar] [CrossRef] [PubMed]
  74. Stergiou, G.; Alpert, B.; Mieke, S.; Wang, J.; O’Brien, E. Validation protocols for blood pressure measuring devices in the 21st century. J. Clin. Hypertens. 2018, 20, 1096–1099. [Google Scholar] [CrossRef] [PubMed]
  75. Stergiou, G.; Alpert, B.; Mieke, S.; Asmar, R.; Atkins, N.; Eckert, S.; Frick, G.; Friedman, B.; Graßl, T.; Ichikawa, T. A universal standard for the validation of blood pressure measuring devices: Association for the Advancement of Medical Instrumentation/European Society of Hypertension/International Organization for Standardization (AAMI/ESH/ISO) Collaboration Statement. Hypertension 2018, 71, 368–374. [Google Scholar] [CrossRef]
  76. Vlachopoulos, C.; O’Rourke, M.; Nichols, W. McDonald’s Blood Flow in Arteries: Theoretical, Experimental and Clinical Principles; CRC Press: Boca Raton, FL, USA, 2011. [Google Scholar]
  77. Giorgini, P.; Weder, A.; Jackson, E.; Brook, R. A review of blood pressure measurement protocols among hypertension trials: Implications for “evidence-based” clinical practice. J. Am. Soc. Hypertens. 2014, 8, 670–676. [Google Scholar] [CrossRef]
  78. Attaei, M.; Khatib, R.; McKee, M.; Lear, S.; Dagenais, G.; Igumbor, E.; AlHabib, K.; Kaur, M.; Kruger, L.; Teo, K. Availability and affordability of blood pressure-lowering medicines and the effect on blood pressure control in high-income, middle-income, and low-income countries: An analysis of the PURE study data. Lancet Public Health 2 2017, 2, e411–e419. [Google Scholar] [CrossRef] [Green Version]
  79. Pickering, T.; Shimbo, D.; Haas, D. Ambulatory blood-pressure monitoring. N. Engl. J. Med. 2006, 354, 2368–2374. [Google Scholar] [CrossRef] [Green Version]
  80. Rosner, B.; Polk, B. Predictive values of routine blood pressure measurements in screening for hypertension. Am. J. Epidemiol. 1983, 117, 429–442. [Google Scholar] [CrossRef]
  81. Sessler, D. Beyond ‘failure to rescue’: The time has come for continuous ward monitoring. Br. J. Anaesth. 2019, 122, 304–306. [Google Scholar] [CrossRef] [Green Version]
  82. Maheshwari, K.; Nathanson, B.; Munson, S.; Khangulov, V.; Stevens, M.; Badani, H.; Khanna, A.; Sessler, D. The relationship between ICU hypotension and in-hospital mortality and morbidity in septic patients. Intensive Care Med. 2018, 44, 857–867. [Google Scholar] [CrossRef] [Green Version]
  83. Zenati, M.; Billiar, T.; Townsend, R.; Peitzman, A.; Harbrecht, B. A brief episode of hypotension increases mortality in critically ill trauma patients. J. Trauma Acute Care Surg. 2002, 53, 232–237. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  84. Taira, D.; Sentell, T.; Albright, C.; Lansidell, D.; Nakagawa, K.; Seto, T.; Stevens, J. Insights in Public Health: Ambulatory Blood Pressure Monitoring: Underuse in Clinical Practice in Hawai’i. Hawaii. J. Med. Public. Health 2017, 76, 314–317. [Google Scholar] [PubMed]
  85. Pickering, T.; Hall, J.; Appel, L.; Falkner, B.; Graves, J.; Hill, M.; Jones, D.; Kurtz, T.; Sheps, S.; Roccella, E.; et al. Recommendations for blood pressure measurement in humans: An AHA scientific statement from the Council on High Blood Pressure Research Professional and Public Education Subcommittee. J. Clin. Hypertens. 2005, 7, 102–109. [Google Scholar] [CrossRef] [PubMed]
  86. Jones, D.; Appel, L.; Sheps, S.; Roccella, E.; Lenfant, C. Measuring blood pressure accurately: New and persistent challenges. JAMA 2003, 289, 1027–1030. [Google Scholar] [CrossRef]
  87. Campbell, N.; Padwal, R.; Picone, D.; Su, H.; Sharman, J. The impact of small to moderate inaccuracies in assessing blood pressure on hypertension prevalence and control rates. J. Clin. Hypertens. 2020, 22, 939–942. [Google Scholar] [CrossRef]
  88. Campbell, N.; McKay, D. Accurate blood pressure measurement: Why does it matter? CMAJ 1999, 161, 277–278. [Google Scholar]
  89. Joffres, M.; Campbell, N.; Manns, B.; Tu, K. Estimate of the benefits of a population-based reduction in dietary sodium additives on hypertension and its related health care costs in Canada. Can. J. Cardiol. 2007, 23, 437–443. [Google Scholar] [CrossRef] [Green Version]
  90. Frieden, T.; Varghese, C.; Kishore, S.; Campbell, N.; Moran, A.; Padwal, R.; Jaffe, M. Scaling up effective treatment of hypertension—A pathfinder for universal health coverage. J. Clin. Hypertens. 2019, 21, 1442–1449. [Google Scholar] [CrossRef] [Green Version]
  91. Parati, G.; Goncalves, A.; Soergel, D.; Bruno, R.; Caiani, E.; Gerdts, E.; Mahfoud, F.; Mantovani, L.; McManus, R.; Santalucia, P. New perspectives for hypertension management: Progress in methodological and technological developments. Eur. J. Prev. Cardiol. 2022. [Google Scholar] [CrossRef]
  92. Padwal, R.; Campbell, N.; Weber, M.; Lackland, D.; Shimbo, D.; Zhang, X.; Schutte, A.; Rakotz, M.; Wozniak, G.; Townsend, R. The Accuracy in Measurement of Blood Pressure (AIM-BP) collaborative: Background and rationale. J. Clin. Hypertens. 2019, 21, 1780–1783. [Google Scholar] [CrossRef] [Green Version]
  93. Campbell, N.; Berbari, A.; Cloutier, L.; Gelfer, M.; Kenerson, J.; Khalsa, T.; Lackland, D.; Lemogoum, D.; Mangat, B.; Mohan, S.; et al. Policy statement of the world hypertension league on noninvasive blood pressure measurement devices and blood pressure measurement in the clinical or community setting. J. Clin. Hypertens. 2014, 16, 320–322. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  94. John, O.; Campbell, N.; Brady, T.; Farrell, M.; Varghese, C.; Berumen, A.V.; Gaitan, V.R.; Laura, A.; Toffelmire, N.; Ameel, M.; et al. The 2020 “WHO technical specifications for automated non-invasive blood pressure measuring devices with cuff”. Hypertension 2021, 77, 806–812. [Google Scholar] [CrossRef] [PubMed]
  95. Picone, D.; Deshpande, R.; Schultz, M.; Fonseca, R.; Campbell, N.; Delles, C.; Olsen, M.H.; Schutte, A.; Stergiou, G.; Padwal, R. Nonvalidated home blood pressure devices dominate the online marketplace in Australia: Major implications for cardiovascular risk management. Hypertension 2020, 75, 1593–1599. [Google Scholar] [CrossRef] [PubMed]
  96. Campbell, N.; Khalsa, T.; Ordunez, P.; Morales, Y.R.; Zhang, X.; Parati, G.; Padwal, R.; Tsuyuki, R.; Cloutier, L.; Sharman, J. Brief online certification course for measuring blood pressure with an automated blood pressure device. A free new resource to support World Hypertension Day Oct 17. J. Clin. Hypertens. 2020, 22, 1754–1756. [Google Scholar]
  97. Zheng, Y.; Ding, X.; Poon, C.; Lo, B.; Zhang, H.; Zhou, X.; Yang, G.; Zhao, N.; Zhang, Y. Unobtrusive sensing and wearable devices for health informatics. IEEE Trans. Biomed. Eng. 2014, 61, 1538–1554. [Google Scholar] [CrossRef]
  98. Luz, E.J.d.S.; Schwartz, W.; Cámara-Chávez, G.; Menotti, D. ECG-based heartbeat classification for arrhythmia detection: A survey. Comput. Methods Programs Biomed. 2016, 127, 144–164. [Google Scholar] [CrossRef]
  99. Estrada, G.M.; Mendoza, L.; Molina, V. Relationship of blood pressure with the electrical signal of the heart using signal processing. Tecciencia 2014, 9, 9–14. [Google Scholar] [CrossRef] [Green Version]
  100. Sharma, M.; Dhiman, H.; Acharya, U. Automatic identification of insomnia using optimal antisymmetric biorthogonal wavelet filter bank with ECG signals. Comput. Biol. Med. 2021, 131, 104246. [Google Scholar] [CrossRef]
  101. Sharma, M.; Agarwal, S.; Acharya, U. Application of an optimal class of antisymmetric wavelet filter banks for obstructive sleep apnea diagnosis using ECG signals. Comput. Biol. Med. 2018, 100, 100–113. [Google Scholar] [CrossRef]
  102. Sharma, M.; Tan, R.S.; Acharya, U. A novel automated diagnostic system for classification of myocardial infarction ECG signals using an optimal biorthogonal filter bank. Comput. Biol. Med. 2018, 102, 341–356. [Google Scholar] [CrossRef]
  103. Reisner, A.; Shaltis, P.; McCombie, D.; Asada, H.; Warner, D.; Warner, M. Utility of the photoplethysmogram in circulatory monitoring. J. Am. Soc. Anesthesiol. 2008, 108, 950–958. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  104. Tamura, T.; Maeda, Y.; Sekine, M.; Yoshida, M. Wearable photoplethysmographic sensors—Past and present. Electronics 2014, 3, 282–302. [Google Scholar] [CrossRef]
  105. Liang, Y.; Chen, Z.; Ward, R.; Elgendi, M. Hypertension assessment using photoplethysmography: A risk stratification approach. J. Clin. Med. 2018, 8, 12. [Google Scholar] [CrossRef] [Green Version]
  106. Inan, O.; Migeotte, P.; Park, K.; Etemadi, M.; Tavakolian, K.; Casanella, R.; Zanetti, J.; Tank, J.; Funtova, I.; Prisk, G. Ballistocardiography and seismocardiography: A review of recent advances. IEEE J. Biomed. Health Inform. 2014, 19, 1414–1427. [Google Scholar] [CrossRef] [Green Version]
  107. Kim, C.; Ober, S.; McMurtry, M.; Finegan, B.; Inan, O.; Mukkamala, R.; Hahn, J. Ballistocardiogram: Mechanism and potential for unobtrusive cardiovascular health monitoring. Sci. Rep. 2016, 6, 31297. [Google Scholar] [CrossRef] [PubMed]
  108. Gill, R. Measurement of blood flow by ultrasound: Accuracy and sources of error. Ultrasound Med. Biol. 1985, 11, 625–641. [Google Scholar] [CrossRef] [PubMed]
  109. Bera, T. Bioelectrical impedance methods for noninvasive health monitoring: A review. J. Med. Eng. 2014, 2014, 381251. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  110. Patterson, R. Fundamentals of impedance cardiography. IEEE Eng. Med. Biol. Mag. 1989, 8, 35–38. [Google Scholar] [CrossRef]
  111. Verkruysse, W.; Svaasand, L.; Nelson, J. Remote plethysmographic imaging using ambient light. Opt. Express 2008, 16, 21434–21445. [Google Scholar] [CrossRef] [Green Version]
  112. Jonathan, E.; Leahy, M. Investigating a smartphone imaging unit for photoplethysmography. Physiol. Meas. 2010, 31, N79. [Google Scholar] [CrossRef]
  113. Landreani, F.; Caiani, E. Smartphone accelerometers for the detection of heart rate. Expert Rev. Med. Devices 2017, 14, 935–948. [Google Scholar] [CrossRef] [PubMed]
  114. Balakrishnan, G.; Durand, F.; Guttag, J. Detecting Pulse from Head Motions in Video. In Proceedings of Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Portland, OR, USA, 23–28 June 2013; pp. 3430–3437. [Google Scholar]
  115. Chung, H.; Rwei, A.; Hourlier-Fargette, A.; Xu, S.; Lee, K.; Dunne, E.; Xie, Z.; Liu, C.; Carlini, A.; Kim, D. Skin-interfaced biosensors for advanced wireless physiological monitoring in neonatal and pediatric intensive-care units. Nat. Med. 2020, 26, 418–429. [Google Scholar] [CrossRef] [PubMed]
  116. Schwartz, G.; Tee, B.; Mei, J.; Appleton, A.; Kim, D.; Wang, H.; Bao, Z. Flexible polymer transistors with high pressure sensitivity for application in electronic skin and health monitoring. Nat. Commun. 2013, 4, 1859. [Google Scholar] [CrossRef] [Green Version]
  117. Wang, C.; Li, X.; Hu, H.; Zhang, L.; Huang, Z.; Lin, M.; Zhang, Z.; Yin, Z.; Huang, B.; Gong, H. Monitoring of the central blood pressure waveform via a conformal ultrasonic device. Nat. Biomed. Eng. 2018, 2, 687–695. [Google Scholar] [CrossRef]
  118. Ha, T.; Tran, J.; Liu, S.; Jang, H.; Jeong, H.; Mitbander, R.; Huh, H.; Qiu, Y.; Duong, J.; Wang, R. A chest-laminated ultrathin and stretchable E-Tattoo for the measurement of electrocardiogram, seismocardiogram, and cardiac time intervals. Adv. Sci. 2019, 6, 1900290. [Google Scholar] [CrossRef]
  119. Di Rienzo, M.; Rizzo, G.; Işilay, Z.; Lombardi, P. SeisMote: A multi-sensor wireless platform for cardiovascular monitoring in laboratory, daily life, and telemedicine. Sensors 2020, 20, 680. [Google Scholar] [CrossRef] [Green Version]
  120. Yano, Y.; Hoshide, S.; Inokuchi, T.; Kanemaru, Y.; Shimada, K.; Kario, K. Association between morning blood pressure surge and cardiovascular remodeling in treated elderly hypertensive subjects. Am. J. Hypertens. 2009, 22, 1177–1182. [Google Scholar] [CrossRef] [Green Version]
  121. Shimizu, M.; Ishikawa, J.; Yano, Y.; Hoshide, S.; Shimada, K.; Kario, K. The relationship between the morning blood pressure surge and low-grade inflammation on silent cerebral infarct and clinical stroke events. Atherosclerosis 2011, 219, 316–321. [Google Scholar] [CrossRef] [PubMed]
  122. Metoki, H.; Ohkubo, T.; Kikuya, M.; Asayama, K.; Obara, T.; Hashimoto, J.; Totsune, K.; Hoshi, H.; Satoh, H.; Imai, Y. Prognostic significance for stroke of a morning pressor surge and a nocturnal blood pressure decline: The Ohasama study. Hypertension 2006, 47, 149–154. [Google Scholar] [CrossRef] [Green Version]
  123. Kuwajima, I.; Mitani, K.; Miyao, M.; Suzuki, Y.; Kuramoto, K.; Ozawa, T. Cardiac implications of the morning surge in blood pressure in elderly hypertensive patients: Relation to arising time. Am. J. Hypertens. 1995, 8, 29–33. [Google Scholar] [CrossRef]
  124. Kario, K.; Pickering, T.; Umeda, Y.; Hoshide, S.; Hoshide, Y.; Morinari, M.; Murata, M.; Kuroda, T.; Schwartz, J.; Shimada, K. Morning surge in blood pressure as a predictor of silent and clinical cerebrovascular disease in elderly hypertensives: A prospective study. Circulation 2003, 107, 1401–1406. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  125. Kaneda, R.; Kario, K.; Hoshide, S.; Umeda, Y.; Hoshide, Y.; Shimada, K. Morning blood pressure hyper-reactivity is an independent predictor for hypertensive cardiac hypertrophy in a community-dwelling population. Am. J. Hypertens. 2005, 18, 1528–1533. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  126. Chen, C.; Li, Y.; Zhang, J.; Wang, Y.; Ling, H.; Chen, K.; Gao, P.; Zhu, D. Association between ambulatory systolic blood pressure during the day and asymptomatic intracranial arterial stenosis. Hypertension 2014, 63, 61–67. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  127. Acciaroli, G.; Facchinetti, A.; Pillonetto, G.; Sparacino, G. Non-Invasive Continuous-Time Blood Pressure Estimation from a Single Channel PPG Signal Using Regularized ARX Models. In Proceedings of the 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Honolulu, HI, USA, 17–21 July 2018; pp. 3630–3633. [Google Scholar]
  128. Kario, K.; Shimada, K. Risers and extreme-dippers of nocturnal blood pressure in hypertension: Antihypertensive strategy for nocturnal blood pressure. Clin. Exp. Hypertens. 2004, 26, 177–189. [Google Scholar] [CrossRef] [PubMed]
  129. Kario, K.; Matsuo, T.; Kobayashi, H.; Imiya, M.; Matsuo, M.; Shimada, K. Nocturnal fall of blood pressure and silent cerebrovascular damage in elderly hypertensive patients: Advanced silent cerebrovascular damage in extreme dippers. Hypertension 1996, 27, 130–135. [Google Scholar] [CrossRef]
  130. Hoshide, S.; Kario, K.; Hoshide, Y.; Umeda, Y.; Hashimoto, T.; Kunii, O.; Ojima, T.; Shimada, K. Associations between nondipping of nocturnal blood pressure decrease and cardiovascular target organ damage in strictly selected community-dwelling normotensives. Am. J. Hypertens. 2003, 16, 434–438. [Google Scholar] [CrossRef] [Green Version]
  131. Burke, L.; Ma, J.; Azar, K.; Bennett, G.; Peterson, E.; Zheng, Y.; Riley, W.; Stephens, J.; Shah, S.; Suffoletto, B. Current science on consumer use of mobile health for cardiovascular disease prevention: A scientific statement from the American Heart Association. Circulation 2015, 132, 1157–1213. [Google Scholar] [CrossRef]
  132. Kuwabara, M.; Harada, K.; Hishiki, Y.; Kario, K. Validation of a wrist-type home nocturnal blood pressure monitor in the sitting and supine position according to the ANSI/AAMI/ISO81060-2: 2013 guidelines: Omron HEM-9600T. J. Clin. Hypertens. 2019, 21, 463–469. [Google Scholar] [CrossRef] [Green Version]
  133. Kuwabara, M.; Harada, K.; Hishiki, Y.; Kario, K. Validation of two watch-type wearable blood pressure monitors according to the ANSI/AAMI/ISO81060-2: 2013 guidelines: Omron HEM-6410T-ZM and HEM-6410T-ZL. J. Clin. Hypertens. 2019, 21, 853–858. [Google Scholar] [CrossRef] [Green Version]
  134. Kollias, A.; Ntineri, A.; Kyriakoulis, K.; Stambolliu, E.; Lagou, S.; Boubouchairopoulou, N.; Stergiou, G. Validation of the professional device for blood pressure measurement Microlife WatchBP Office in adults and children according to the American National Standards Institute/Association for the Advancement of Medical Instrumentation/International Organization for Standardization standard. Blood Press. Monit. 2018, 23, 112–114. [Google Scholar]
  135. Kario, K.; Shimbo, D.; Tomitani, N.; Kanegae, H.; Schwartz, J.; Williams, B. The first study comparing a wearable watch-type blood pressure monitor with a conventional ambulatory blood pressure monitor on in-office and out-of-office settings. J. Clin. Hypertens. 2020, 22, 135–141. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  136. Islam, S.; Cartledge, S.; Karmakar, C.; Rawstorn, J.; Fraser, S.; Chow, C.; Maddison, R. Validation and acceptability of a cuffless wrist-worn wearable blood pressure monitoring device among users and health care professionals: Mixed methods study. JMIR mHealth uHealth 2019, 7, e14706. [Google Scholar] [CrossRef] [PubMed]
  137. Watanabe, N.; Bando, Y.; Kawachi, T.; Yamakita, H.; Futatsuyama, K.; Honda, Y.; Yasui, H.; Nishimura, K.; Kamihara, T.; Okumura, T. Development and validation of a novel cuff-less blood pressure monitoring device. Basic Transl. Sci. 2017, 2, 631–642. [Google Scholar] [CrossRef] [PubMed]
  138. Lazazzera, R.; Belhaj, Y.; Carrault, G. A new wearable device for blood pressure estimation using photoplethysmogram. Sensors 2019, 19, 2557. [Google Scholar] [CrossRef] [Green Version]
  139. Chan, G.; Cooper, R.; Hosanee, M.; Welykholowa, K.; Kyriacou, P.; Zheng, D.; Allen, J.; Abbott, D.; Lovell, N.; Fletcher, R. Multi-site photoplethysmography technology for blood pressure assessment: Challenges and recommendations. J. Clin. Med. 2019, 8, 1827. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  140. Bilo, G.; Zorzi, C.; Munera, J.O.; Torlasco, C.; Giuli, V.; Parati, G. Validation of the Somnotouch-NIBP noninvasive continuous blood pressure monitor according to the European Society of Hypertension International Protocol revision. Blood Press. Monit. 2015, 20, 291–294. [Google Scholar] [CrossRef]
  141. O’Brien, E.; Pickering, T.; Asmar, R.; Myers, M.; Parati, G.; Staessen, J.; Mengden, T.; Imai, Y.; Waeber, B.; Palatini, P. Working Group on Blood Pressure Monitoring of the European Society of Hypertension International Protocol for validation of blood pressure measuring devices in adults. Blood Press. Monit. 2002, 7, 3–17. [Google Scholar] [CrossRef]
  142. Ibrahim, B.; Jafari, R. Continuous Blood Pressure Monitoring Using Wrist-Worn Bio-Impedance Sensors with Wet Electrodes. In Proceedings of the IEEE Biomedical Circuits and Systems Conference (BioCAS), Cleveland, OH, USA, 17–19 October 2018; pp. 1–4. [Google Scholar]
  143. Qiu, C.; Wu, T.; Redouté, J.; Yuce, M. A Wireless Wearable Sensor Patch for the Real-Time Estimation of Continuous Beat-to-Beat Blood Pressure. In Proceedings of the 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Berlin, Germany, 23–27 July 2019; pp. 6842–6845. [Google Scholar]
  144. Hosanee, M.; Chan, G.; Welykholowa, K.; Cooper, R.; Kyriacou, P.; Zheng, D.; Allen, J.; Abbott, D.; Menon, C.; Lovell, N. Cuffless single-site photoplethysmography for blood pressure monitoring. J. Clin. Med. 2020, 9, 723. [Google Scholar] [CrossRef] [Green Version]
  145. Lee, H.; Lee, D.; Seo, J.; Ihm, S.; Kim, K.; Cho, E.; Kim, H.; Shin, J.; Park, S.; Sohn, I. Smartphone/smartwatch-based cuffless blood pressure measurement: A position paper from the Korean Society of Hypertension. Clin. Hypertens. 2021, 27, 1–8. [Google Scholar] [CrossRef]
  146. Arakawa, T. Recent research and developing trends of wearable sensors for detecting blood pressure. Sensors 2018, 18, 2772. [Google Scholar] [CrossRef] [Green Version]
  147. Boubouchairopoulou, N.; Kollias, A.; Chiu, B.; Chen, B.; Lagou, S.; Anestis, P.; Stergiou, G. A novel cuffless device for self-measurement of blood pressure: Concept, performance and clinical validation. J. Hum. Hypertens. 2017, 31, 479–482. [Google Scholar] [CrossRef] [PubMed]
  148. Mamun; Khan, M.M.R. Significance of Features from Biomedical Signals in Heart Health Monitoring. BioMed 2022, 2, 391–408. [Google Scholar] [CrossRef]
  149. Vybornova, A.; Polychronopoulou, E.; Wurzner-Ghajarzadeh, A.; Fallet, S.; Sola, J.; Wuerzner, G. Blood pressure from the optical Aktiia Bracelet: A 1-month validation study using an extended ISO81060-2 protocol adapted for a cuffless wrist device. Blood Press. Monit. 2021, 26, 305–311. [Google Scholar] [CrossRef] [PubMed]
  150. Kenney, M.; Seals, D. Postexercise hypotension. Key features, mechanisms, and clinical significance. Hypertension 1993, 22, 653–664. [Google Scholar] [CrossRef] [Green Version]
  151. Mukkamala, R.; Hahn, J. Initialization of Pulse Transit Time-Based Blood Pressure Monitors. In The Handbook of Cuffless Blood Pressure Monitoring; Springer: Berlin/Heidelberg, Germany, 2019; pp. 163–190. [Google Scholar]
  152. Miao, F.; Liu, Z.; Liu, J.; Wen, B.; He, Q.; Li, Y. Multi-sensor fusion approach for cuff-less blood pressure measurement. IEEE J. Biomed. Health Inform. 2019, 24, 79–91. [Google Scholar] [CrossRef]
  153. Liu, Z.; Zhou, B.; Li, Y.; Tang, M.; Miao, F. Continuous blood pressure estimation from electrocardiogram and photoplethysmogram during arrhythmias. Front. Physiol. 2020, 11, 575407. [Google Scholar] [CrossRef]
  154. Xing, X.; Ma, Z.; Zhang, M.; Zhou, Y.; Dong, W.; Song, M. An unobtrusive and calibration-free blood pressure estimation method using photoplethysmography and biometrics. Sci. Rep. 2019, 9, 8611 . [Google Scholar] [CrossRef] [Green Version]
  155. Poon, C.; Zhang, Y. Cuff-Less and Noninvasive Measurements of Arterial Blood Pressure by Pulse Transit Time. In Proceedings of the IEEE Engineering in Medicine and Biology 27th Annual Conference, Shanghai, China, 17–18 January 2006; pp. 5877–5880. [Google Scholar]
  156. Schoettker, P.; Degott, J.; Hofmann, G.; Proença, M.; Bonnier, G.; Lemkaddem, A.; Lemay, M.; Schorer, R.; Christen, U.; Knebel, J. Blood pressure measurements with the OptiBP smartphone app validated against reference auscultatory measurements. Sci. Rep. 2020, 10, 17827. [Google Scholar] [CrossRef]
  157. Ruiz-Rodríguez, J.; Ruiz-Sanmartín, A.; Ribas, V.; Caballero, J.; García-Roche, A.; Riera, J.; Nuvials, X.; de Nadal, M.; de Sola-Morales, O.; Serra, J. Innovative continuous non-invasive cuffless blood pressure monitoring based on photoplethysmography technology. Intensive Care Med. 2013, 39, 1618–1625. [Google Scholar] [CrossRef]
  158. Luo, H.; Yang, D.; Barszczyk, A.; Vempala, N.; Wei, J.; Wu, S.; Zheng, P.; Fu, G.; Lee, K.; Feng, Z. Smartphone-based blood pressure measurement using transdermal optical imaging technology. Circ. Cardiovasc. Imaging 2019, 12, e008857. [Google Scholar] [CrossRef] [Green Version]
  159. Nachman, D.; Gepner, Y.; Goldstein, N.; Kabakov, E.; Ishay, A.; Littman, R.; Azmon, Y.; Jaffe, E.; Eisenkraft, A. Comparing blood pressure measurements between a photoplethysmography-based and a standard cuff-based manometry device. Sci. Rep. 2020, 10, 16116. [Google Scholar] [CrossRef] [PubMed]
  160. Allen, J. Photoplethysmography and its application in clinical physiological measurement. Physiol. Meas. 2007, 28, R1. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  161. Harma, M.; Barbosa, K.; Ho, V.; Griggs, D.; Ghirmai, T.; Krishnan, S.; Hsiai, T.; Chiao, J.; Cao, H. Cuff-less and continuous blood pressure monitoring: A methodological review. Technologies 2017, 5, 21. [Google Scholar]
  162. Gao, M.; Olivier, N.; Mukkamala, R. Comparison of noninvasive pulse transit time estimates as markers of blood pressure using invasive pulse transit time measurements as a reference. Physiol. Rep. 2016, 4, e12768. [Google Scholar] [CrossRef]
  163. Allen, J.; Murray, A. Age-related changes in peripheral pulse timing characteristics at the ears, fingers and toes. J. Hum. Hypertens. 2002, 16, 711–717. [Google Scholar] [CrossRef] [Green Version]
  164. Chen, Y.; Wen, C.; Tao, G.; Bi, M. Continuous and noninvasive measurement of systolic and diastolic blood pressure by one mathematical model with the same model parameters and two separate pulse wave velocities. Ann. Biomed. Eng. 2012, 40, 871–882. [Google Scholar] [CrossRef]
  165. Mamun, K.; Rahman, M.M.; Alouani, A. Using Photoplethysmography & ECG towards a Non-Invasive Cuff Less Blood Pressure Measurement Technique. In Proceedings of the 2019 IEEE Canadian Conference of Electrical and Computer Engineering (CCECE), Edmonton, AB, Canada, 5–8 May 2019; pp. 1–4. [Google Scholar]
  166. Nitzan, M.; Khanokh, B.; Slovik, Y. The difference in pulse transit time to the toe and finger measured by photoplethysmography. Physiol. Meas. 2001, 23, 85. [Google Scholar] [CrossRef]
  167. Liang, Y.; Abbott, D.; Howard, N.; Lim, K.; Ward, R.; Elgendi, M. How effective is pulse arrival time for evaluating blood pressure? Challenges and recommendations from a study using the MIMIC database. J. Clin. Med. 2019, 8, 337. [Google Scholar] [CrossRef] [Green Version]
  168. Kim, C.; Carek, A.; Inan, O.; Mukkamala, R.; Hahn, J. Ballistocardiogram-based approach to cuffless blood pressure monitoring: Proof of concept and potential challenges. IEEE Trans. Biomed. Eng. 2018, 65, 2384–2391. [Google Scholar] [CrossRef]
  169. Lee, J.; Yang, S.; Lee, S.; Kim, H. Analysis of pulse arrival time as an indicator of blood pressure in a large surgical biosignal database: Recommendations for developing ubiquitous blood pressure monitoring methods. J. Clin. Med. 2019, 8, 1773. [Google Scholar]
  170. Sharwood-Smith, G.; Bruce, J. Drummond. Assessment of pulse transit time to indicate cardiovascular changes during obstetric spinal anaesthesia. Br. J. Anaesth. 2006, 96, 100–105. [Google Scholar] [CrossRef] [Green Version]
  171. Yoon, Y.; Kang, J.; Kwon, Y.; Park, S.; Noh, S.; Kim, Y.; Park, J.; Hwang, S. Cuff-less blood pressure estimation using pulse waveform analysis and pulse arrival time. IEEE J. Biomed. Health Inform. 2017, 22, 1068–1074. [Google Scholar] [CrossRef]
  172. Noordergraaf, A. Circulatory System Dynamics; Elsevier: Amsterdam, The Netherlands, 2012. [Google Scholar]
  173. Zhang, G.; Gao, M.; Xu, D.; Olivier, N.; Mukkamala, R. Pulse arrival time is not an adequate surrogate for pulse transit time as a marker of blood pressure. J. Appl. Physiol. 2011, 111, 1681–1686. [Google Scholar] [CrossRef] [Green Version]
  174. Chen, W.; Kobayashi, T.; Ichikawa, S.; Takeuchi, Y.; Togawa, T. Continuous estimation of systolic blood pressure using the pulse arrival time and intermittent calibration. Med. Biol. Eng. Comput. 2000, 38, 569–574. [Google Scholar] [CrossRef]
  175. Forouzanfar, M.; Ahmad, S.; Batkin, I.; Dajani, H.; Groza, V.; Bolic, M. Model-based mean arterial pressure estimation using simultaneous electrocardiogram and oscillometric blood pressure measurements. IEEE Trans. Instrum. Meas. 2015, 64, 2443–2452. [Google Scholar] [CrossRef]
  176. Ma, T.; Zhang, Y. A Correlation Study on the Variabilities in Pulse Transit Time, Blood Pressure, and Heart Rate Recorded Simultaneously from Healthy Subjects. In Proceedings of the 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference, Shanghai, China, 17–18 January 2006; pp. 996–999. [Google Scholar]
  177. Sigrist, R.; Liau, J.; Kaffas, A.; Chammas, M.; Willmann, J. Ultrasound Elastography: Review of Techniques and Clinical Applications. Theranostics 2017, 7, 1303–1329. [Google Scholar] [CrossRef]
  178. Li, J.; Li, R.; Chen, Z.; Deng, G.; Wang, H.; Mavromoustakis, C.; Song, H.; Ming, Z. Design of a continuous blood pressure measurement system based on pulse wave and ECG signals. IEEE J. Transl. Eng. Health Med. 2018, 6, 1–14. [Google Scholar] [CrossRef]
  179. Byfield, R.; Miller, M.; Miles, J.; Guidoboni, G.; Lin, J. Towards Robust Blood Pressure Estimation From Pulse Wave Velocity Measured by Photoplethysmography Sensors. IEEE Sens. J. 2021, 22, 2475–2483. [Google Scholar] [CrossRef]
  180. Kao, Y.; Chao, P.; Wey, C. Design and validation of a new PPG module to acquire high-quality physiological signals for high-accuracy biomedical sensing. IEEE J. Sel. Top. Quantum Electron. 2018, 25, 1–10. [Google Scholar] [CrossRef]
  181. Kachuee, M.; Kiani, M.; Mohammadzade, H.; Shabany, M. Cuffless blood pressure estimation algorithms for continuous health-care monitoring. IEEE Trans. Biomed. Eng. 2016, 64, 859–869. [Google Scholar] [CrossRef]
  182. Kachuee, M.; Kiani, M.; Mohammadzade, H.; Shabany, M. Cuff-Less High-Accuracy Calibration-Free Blood Pressure Estimation Using Pulse Transit Time. In Proceedings of the 2015 IEEE International Symposium on Circuits and Systems (ISCAS), Lisbon, Portugal, 24–27 May 2015; pp. 1006–1009. [Google Scholar]
  183. Goldberger, A.; Amaral, L.; Glass, L.; Hausdorff, J.; Ivanov, P.; Mark, R.; Mietus, J.; Moody, G.; Peng, C.; Stanley, H. PhysioBank, PhysioToolkit, and PhysioNet: Components of a new research resource for complex physiologic signals. Circulation 2000, 101, e215–e220. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  184. Suzuki, S.; Oguri, K. Cuffless Blood Pressure Estimation by Error-Correcting Output Coding Method Based on an Aggregation of AdaBoost with a Photoplethysmograph Sensor. In Proceedings of the 2009 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Minneapolis, MN, USA, 3–6 September 2009; pp. 6765–6768. [Google Scholar]
  185. Teng, X.; Zhang, Y. Continuous and noninvasive estimation of arterial blood pressure using a photoplethysmographic approach. In Proceedings of the 25th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (IEEE Cat. No. 03CH37439), Cancun, Mexico, 17–21 September 2003; 4, pp. 3153–3156. [Google Scholar]
  186. Peter, L.; Kracik, J.; Cerny, M.; Noury, N.; Polzer, S. Mathematical model based on the shape of pulse waves measured at a single spot for the non-invasive prediction of blood pressure. Processes 2020, 8, 442. [Google Scholar] [CrossRef] [Green Version]
  187. Monte-Moreno, E. Non-invasive estimate of blood glucose and blood pressure from a photoplethysmograph by means of machine learning techniques. Artif. Intell. Med. 2011, 53, 127–138. [Google Scholar] [CrossRef]
  188. Kavsaoğlu, A.; Polat, K.; Hariharan, M. Non-invasive prediction of hemoglobin level using machine learning techniques with the PPG signal’s characteristics features. Appl. Soft Comput. 2015, 37, 983–991. [Google Scholar] [CrossRef]
  189. Elgendi, M. On the analysis of fingertip photoplethysmogram signals. Curr. Cardiol. Rev. 2012, 8, 14–25. [Google Scholar] [CrossRef] [PubMed]
  190. Mahbub, Z.; Rabbani, K. Frequency domain analysis to identify neurological disorders from evoked EMG responses. J. Biol. Phys. 2007, 33, 99–108. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  191. Zhang, Y.; Feng, Z. A SVM Method for Continuous Blood Pressure Estimation from a PPG Signal. In Proceedings of the 9th International Conference on Machine Learning and Computing, New York, NY, USA, 24–26 February 2017; pp. 128–132. [Google Scholar]
  192. Xing, X.; Sun, M. Optical blood pressure estimation with photoplethysmography and FFT-based neural networks. Biomed. Opt. Express 2016, 7, 3007–3020. [Google Scholar] [CrossRef] [Green Version]
  193. Cattivelli, F.; Garudadri, H. Noninvasive Cuffless Estimation of Blood Pressure from Pulse Arrival Time and Heart Rate with Adaptive Calibration. In Proceedings of the 2009 Sixth International Workshop on Wearable and Implantable Body Sensor Networks, Berkeley, CA, USA, 3–5 June 2009; pp. 114–119. [Google Scholar]
  194. Chowdhury, M.; Khandakar, A.; Alzoubi, K.; Mansoor, S.; Tahir, A.M.; Reaz, M.; Al-Emadi, N. Real-time smart-digital stethoscope system for heart diseases monitoring. Sensors 2019, 19, 2781. [Google Scholar] [CrossRef]
  195. Chowdhury, M.; Alzoubi, K.; Khandakar, A.; Khallifa, R.; Abouhasera, R.; Koubaa, S.; Ahmed, R.; Hasan, A. Wearable real-time heart attack detection and warning system to reduce road accidents. Sensors 2019, 19, 2780. [Google Scholar] [CrossRef] [Green Version]
  196. Yi, C.; Jian, C.; Wenqiang, J. Continuous Blood Pressure Measurement Based on Photoplethysmography. In Proceedings of the 14th IEEE International Conference on Electronic Measurement & Instruments (ICEMI), Changsha, China, 1–3 November 2019; pp. 1656–1663. [Google Scholar]
  197. Miao, F.; Fu, N.; Zhang, Y.; Ding, X.; Hong, X.; He, Q.; Li, Y. A novel continuous blood pressure estimation approach based on data mining techniques. IEEE J. Biomed. Health Inform. 2017, 21, 1730–1740. [Google Scholar] [CrossRef] [PubMed]
  198. Hsieh, Y.; Wu, C.; Lu, S.; Tsao, Y. A Linear Regression Model with Dynamic Pulse Transit Time Features for Noninvasive Blood Pressure Prediction. In Proceedings of the 2016 IEEE Biomedical Circuits and Systems Conference (BioCAS), Shanghai, China, 17–19 October 2016; pp. 604–607. [Google Scholar]
  199. Ghosh, S.; Banerjee, A.; Ray, N.; Wood, P.; Boulanger, P.; Padwal, R. Continuous Blood Pressure Prediction from Pulse Transit Time Using ECG and PPG Signals. In Proceedings of the 2016 IEEE Healthcare Innovation Point-Of-Care Technologies Conference (HI-POCT), Cancun, Mexico, 9–11 November 2016; pp. 188–191. [Google Scholar]
  200. Datta, S.; Banerjee, R.; Choudhury, A.; Sinha, A.; Pal, A. Blood Pressure Estimation from Photoplethysmogram Using Latent Parameters. In Proceedings of the 2016 IEEE International Conference on Communications (ICC), Kuala Lumpur, Malaysia, 22–27 May 2016; pp. 1–7. [Google Scholar]
  201. Choudhury, A.; Banerjee, R.; Sinha, A.; Kundu, S. Estimating Blood Pressure Using Windkessel Model on Photoplethysmogram. In Proceedings of the 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Chicago, IL, USA, 26–30 August 2014; pp. 4567–4570. [Google Scholar]
  202. Wong, M.; Poon, C.; Zhang, Y. An evaluation of the cuffless blood pressure estimation based on pulse transit time technique: A half year study on normotensive subjects. Cardiovasc. Eng. 2009, 9, 32–38. [Google Scholar] [CrossRef] [PubMed]
  203. Hassan; Ali, M.K.B.; Mashor, M.; Nasir, N.; Mohamed, S. Measuring Blood Pressure Using a Photoplethysmography Approach. In 4th Kuala Lumpur International Conference on Biomedical Engineering; Springer: Berlin/Heidelberg, Germany, 2008; pp. 591–594. [Google Scholar]
  204. Gao, S.; Wittek, P.; Zhao, L.; Jiang, W. Data-Driven Estimation of Blood Pressure Using Photoplethysmographic Signals. In Proceedings of the 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Orlando, FL, USA, 16–20 August 2016; pp. 766–769. [Google Scholar]
  205. Liu, M.; Po, L.; Fu, H. Cuffless blood pressure estimation based on photoplethysmography signal and its second derivative. Int. J. Comput. Theory Eng. 2017, 9, 202. [Google Scholar] [CrossRef] [Green Version]
  206. Khalid, S.; Zhang, J.; Chen, F.; Zheng, D. Blood pressure estimation using photoplethysmography only: Comparison between different machine learning approaches. J. Healthc. Eng. 2018, 2018, 1548647. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  207. Zhang, B.; Ren, H.; Huang, G.; Cheng, Y.; Hu, C. Predicting blood pressure from physiological index data using the SVR algorithm. BMC Bioinform. 2019, 20, 1–15. [Google Scholar] [CrossRef] [PubMed]
  208. Dey, J.; Gaurav, A.; Tiwari, V. InstaBP: Cuff-Less Blood Pressure Monitoring on Smartphone Using Single PPG Sensor. In Proceedings of the 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Honolulu, HI, USA, 18–21 July 2018; pp. 5002–5005. [Google Scholar]
  209. Fujita, D.; Suzuki, A.; Ryu, K. PPG-based systolic blood pressure estimation method using PLS and level-crossing feature. Appl. Sci. 2019, 9, 304. [Google Scholar] [CrossRef] [Green Version]
  210. Mousavi, S.; Firouzmand, M.; Charmi, M.; Hemmati, M.; Moghadam, M.; Ghorbani, Y. Blood pressure estimation from appropriate and inappropriate PPG signals using A whole-based method. Biomed. Signal Process. Control. 2019, 47, 196–206. [Google Scholar] [CrossRef]
  211. Mamun, K.; Rahman, M.M.; Alouani, A. Cuffless Blood Pressure Measurement Using Linear and Nonlinear Optimized Feature Selection. Diagnostics 2022, 12, 408. [Google Scholar] [CrossRef]
  212. Landry, C.; Peterson, S.; Arami, A. Nonlinear dynamic modeling of blood pressure waveform: Towards an accurate cuffless monitoring system. IEEE Sens. J. 2020, 20, 5368–5378. [Google Scholar] [CrossRef]
  213. Mehrabadi, M.; Aqajari, S.; Zargari, A.; Dutt, N.; Rahmani, A. Novel Blood Pressure Waveform Reconstruction from Photoplethysmography using Cycle Generative Adversarial Networks. arXiv 2022, arXiv:2201.09976. [Google Scholar]
  214. Lamonaca, F.; Barbe, K.; Kurylyak, Y.; Grimaldi, D.; Van Moer, W.; Furfaro, A.; Spagnuolo, V. Application of the Artificial Neural Network for Blood Pressure Evaluation with Smartphones. In Proceedings of the 2013 IEEE 7th International Conference on Intelligent Data Acquisition and Advanced Computing Systems (IDAACS), Berlin, Germany, 12–14 September 2013; pp. 408–412. [Google Scholar]
  215. Su, P.; Ding, X.; Zhang, Y.; Liu, J.; Miao, F.; Zhao, N. Long-Term Blood Pressure Prediction with Deep Recurrent Neural Networks. In Proceedings of the 2018 IEEE EMBS International Conference on Biomedical & Health Informatics (BHI), Las Vegas, NV, USA, 4 March 2018; pp. 323–328. [Google Scholar]
  216. Liang, Y.; Chen, Z.; Ward, R.; Elgendi, M. Photoplethysmography and deep learning: Enhancing hypertension risk stratification. Biosensors 2018, 8, 101. [Google Scholar] [PubMed] [Green Version]
  217. Panwar, M.; Gautam, A.; Biswas, D.; Acharyya, A. PP-Net: A deep learning framework for PPG-based blood pressure and heart rate estimation. IEEE Sensors J. 2020, 20, 10000–10011. [Google Scholar] [CrossRef]
  218. Tazarv, A.; Levorato, M. A Deep Learning Approach to Predict Blood Pressure from PPG Signals. In Proceedings of the 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), Guadalajara, Mexico, 31 October–4 November 2021; pp. 5658–5662. [Google Scholar]
  219. Jeong, D.; Lim, K. Combined deep CNN–LSTM network-based multitasking learning architecture for noninvasive continuous blood pressure estimation using difference in ECG-PPG features. Sci. Rep. 2021, 11, 13539. [Google Scholar] [CrossRef]
  220. Schrumpf, F.; Frenzel, P.; Aust, C.; Osterhoff, G.; Fuchs, M. Assessment of non-invasive blood pressure prediction from ppg and rppg signals using deep learning. Sensors 2021, 21, 6022. [Google Scholar] [CrossRef] [PubMed]
  221. Harfiya, L.; Chang, C.; Li, Y. Continuous blood pressure estimation using exclusively photopletysmography by LSTM-based signal-to-signal translation. Sensors 2021, 21, 2952. [Google Scholar] [CrossRef]
  222. Huang, B.; Chen, W.; Lin, C.; Juang, C.; Wang, J. MLP-BP: A novel framework for cuffless blood pressure measurement with PPG and ECG signals based on MLP-Mixer neural networks. Biomed. Signal Process. Control 2022, 73, 103404. [Google Scholar] [CrossRef]
  223. Wu, B.; Chiu, L.; Wu, Y.; Lai, C.; Chu, P. Contactless Blood Pressure Measurement via Remote Photoplethysmography with Synthetic Data Generation Using Generative Adversarial Network. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA, 19–20 June 2022; pp. 2130–2138. [Google Scholar]
  224. Mamun, K.; Rahman, M.M.; Alouani, A. FA-1D-CNN Implementation to Improve Diagnosis of Heart Disease Risk Level. In Proceedings of the 6th World Congress on Electrical Engineering and Computer Systems and Science, Prague, Czech Republic, 13–15 August 2020; pp. 122.1–122.9. [Google Scholar]
  225. Wu, D.; Xu, L.; Zhang, R.; Zhang, H.; Ren, L.; Zhang, Y. Continuous cuff-less blood pressure estimation based on combined information using deep learning approach. J. Med. Imaging Health Inform. 2018, 8, 1290–1299. [Google Scholar]
  226. Vardhan, K.; Vedanth, S.; Poojah, G.; Abhishek, K.; Kumar, M.; Vijayaraghavan, V. BP-Net: Efficient Deep Learning for Continuous Arterial Blood Pressure Estimation using Photoplethysmogram. In Proceedings of the 2021 20th IEEE International Conference on Machine Learning and Applications (ICMLA), Pasadena, CA, USA, 13–16 December 2021; pp. 1495–1500. [Google Scholar]
  227. Liu, D.; Gorges, M.; Jenkins, S. University of Queensland vital signs dataset: Development of an accessible repository of anesthesia patient monitoring data for research. Anesth. Analg. 2012, 114, 584–589. [Google Scholar] [CrossRef]
  228. Johnson, A.; Pollard, T.; Shen, L.; Lehman, L.; Feng, M.; Ghassemi, M.; Moody, B.; Szolovits, P.; Celi, L.A.; Mark, R. MIMIC-III, a freely accessible critical care database. Sci. Data 2016, 3, 1–9. [Google Scholar] [CrossRef] [Green Version]
  229. Saeed, M.; Villarroel, M.; Reisner, A.; Clifford, G.; Lehman, L.; Moody, G.; Heldt, T.; Kyaw, T.; Moody, B.; Mark, R. Multiparameter Intelligent Monitoring in Intensive Care II: A public-access intensive care unit database. Crit. Care Med. 2011, 39, 952–960. [Google Scholar] [CrossRef] [Green Version]
  230. Schlesinger, O.; Vigderhouse, N.; Eytan, D.; Moshe, Y. Blood Pressure Estimation from PPG Signals Using Convolutional Neural Networks and Siamese Network. In Proceedings of the ICASSP 2020–2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Barcelona, Spain, 4–8 May 2020; pp. 1135–1139. [Google Scholar]
  231. Liang, Y.; Chen, Z.; Ward, R.; Elgendi, M. Hypertension assessment via ECG and PPG signals: An evaluation using MIMIC database. Diagnostics 2018, 8, 65. [Google Scholar] [CrossRef] [PubMed]
  232. Melillo, P.; Izzo, R.; Orrico, A.; Scala, P.; Attanasio, M.; Mirra, M.; De Luca, N.; Pecchia, L. Automatic prediction of cardiovascular and cerebrovascular events using heart rate variability analysis. PloS ONE 2015, 10, e0118504. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  233. Jain, P.; Gajbhiye, P.; Tripathy, R.; Acharya, U. A two-stage Deep CNN Architecture for the Classification of Low-risk and High-risk Hypertension Classes using Multi-lead ECG Signals. Inform. Med. Unlocked 2020, 21, 100479. [Google Scholar] [CrossRef]
  234. Ding, X.; Yan, B.; Zhang, Y.; Liu, J.; Zhao, N.; Tsang, H. Pulse transit time based continuous cuffless blood pressure estimation: A new extension and a comprehensive evaluation. Sci. Rep. 2017, 7, 11554. [Google Scholar] [CrossRef] [PubMed]
  235. Huynh, T.; Jafari, R.; Chung, W. Noninvasive cuffless blood pressure estimation using pulse transit time and impedance plethysmography. IEEE Trans. Biomed. Eng. 2018, 66, 967–976. [Google Scholar] [CrossRef]
  236. Seals, S.; Colantonio, L.; Tingle, J.; Shimbo, D.; Correa, A.; Griswold, M.; Muntner, P. Calibration of blood pressure measurements in the Jackson Heart Study. Blood Press. Monit. 2019, 24, 130–136. [Google Scholar] [CrossRef]
Figure 1. Flow diagram of the article selection process for literature review.
Figure 1. Flow diagram of the article selection process for literature review.
Bioengineering 10 00027 g001
Figure 2. Traditional methods of BP measurement techniques. (a) arterial catheterization method, (b) vascular unloading technique, (c) arterial tonometry method, (d) auscultation method, (e) oscillometric method [19,43] CC by 4.0.
Figure 2. Traditional methods of BP measurement techniques. (a) arterial catheterization method, (b) vascular unloading technique, (c) arterial tonometry method, (d) auscultation method, (e) oscillometric method [19,43] CC by 4.0.
Bioengineering 10 00027 g002
Figure 3. Technological advances behind the research progress in cuffless BP measurement techniques.
Figure 3. Technological advances behind the research progress in cuffless BP measurement techniques.
Bioengineering 10 00027 g003
Figure 4. Biomedical signals used for recent BP measurement techniques (noninvasive and cuffless) [119] CC 4.0.
Figure 4. Biomedical signals used for recent BP measurement techniques (noninvasive and cuffless) [119] CC 4.0.
Bioengineering 10 00027 g004
Figure 5. Example of pulse transit time [167] CC by 4.0.
Figure 5. Example of pulse transit time [167] CC by 4.0.
Bioengineering 10 00027 g005
Figure 6. Typical steps for pulse wave analysis to get BP measurement model.
Figure 6. Typical steps for pulse wave analysis to get BP measurement model.
Bioengineering 10 00027 g006
Figure 7. Example of features in pulse wave analysis [186] CC by 4.0.
Figure 7. Example of features in pulse wave analysis [186] CC by 4.0.
Bioengineering 10 00027 g007
Figure 8. A summary of features used from biomedical signals (typically PPG wave) to create a model for BP measurement (cuffless-noninvasive).
Figure 8. A summary of features used from biomedical signals (typically PPG wave) to create a model for BP measurement (cuffless-noninvasive).
Bioengineering 10 00027 g008
Figure 9. An example of a deep learning network flow diagram.
Figure 9. An example of a deep learning network flow diagram.
Bioengineering 10 00027 g009
Table 1. Datasets used in different BP measurement models.
Table 1. Datasets used in different BP measurement models.
DatasetDataSignalDescriptionReference Research
MIMIC II [229]26,870PPGIt contains different physiological signals, including blood pressure, PPG, and ECG. There is confusion about whether the data were synchronized, so using MIMIC II for any data such as PAT is not recommended.[230]
MIMIC III [228]40,000ECG, PPGIt is a collection of physiological data from a different hospital; most research with clinical data for the BP measurement model has used this dataset.[231]
University of Queensland’s vital signs dataset [227]23,617PPGThis dataset includes blood pressure measurement along with PPG signal. The case number is limited compared to MIMIC II and III.[218]
SHAREE (Smart Health for Assessing the Risk of Events via ECG) [232]24 h electrocardiographic (ECG) Holter recordings of 139 hypertensive patientsECGAnti-hypertensive treatment was given to 139 patients, and after one month, the ECG Holter was used to record data. Each recording was of 24 h, containing three ECG leads sampling at 128 Hz.[233]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Khan Mamun, M.M.R.; Sherif, A. Advancement in the Cuffless and Noninvasive Measurement of Blood Pressure: A Review of the Literature and Open Challenges. Bioengineering 2023, 10, 27.

AMA Style

Khan Mamun MMR, Sherif A. Advancement in the Cuffless and Noninvasive Measurement of Blood Pressure: A Review of the Literature and Open Challenges. Bioengineering. 2023; 10(1):27.

Chicago/Turabian Style

Khan Mamun, Mohammad Mahbubur Rahman, and Ahmed Sherif. 2023. "Advancement in the Cuffless and Noninvasive Measurement of Blood Pressure: A Review of the Literature and Open Challenges" Bioengineering 10, no. 1: 27.

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop