Sensors

Research

12 pages, 3297 KiB

Open AccessArticle

Electrodermal Activity Analysis at Different Body Locations

by Patricia Gamboa, Rui Varandas, Katrin Mrotzeck, Hugo Plácido da Silva and Cláudia Quaresma

Sensors 2025, 25(6), 1762; https://doi.org/10.3390/s25061762 - 12 Mar 2025

Viewed by 562

Electrodermal activity (EDA) reflects the variation in the electrical conductance of the skin in response to sweat secretion, constituting a non-invasive measure of the sympathetic nervous system. This system intervenes in reactions to stress and is strongly activated in emotional states. In most [...] Read more.

Electrodermal activity (EDA) reflects the variation in the electrical conductance of the skin in response to sweat secretion, constituting a non-invasive measure of the sympathetic nervous system. This system intervenes in reactions to stress and is strongly activated in emotional states. In most cases, EDA signals are collected from the hand (fingers or palms), which is not an ideal location for a sensor when the participant has to use their hands during tasks or activities. This study aims to explore alternative locations for retrieving EDA signals (e.g., the chest, back, and forehead). EDA signals from 25 healthy participants were collected using a protocol involving different physical stimuli that have been reported to induce an electrodermal response. The features extracted included the Skin Conductance Response (SCR) height, SCR amplitude, and peak prominence. An analysis of these features and the analysis of the correlation between the standard position with the different locations suggested that the chest, while a possible alternative for EDA signal collection, presents some weak results, and further evaluation of this site is needed. Additionally, the forehead should be excluded as an alternative site, at least in short-term measurements. Full article

(This article belongs to the Special Issue Sensors for Physiological Monitoring and Digital Health)

► Show Figures

Figure 1

19 pages, 2159 KiB

Open AccessArticle

Correlation Between Pain Intensity and Trunk Sway in Seated Posture Among Office Workers with Chronic Spinal Pain: A Pilot Field-Based Study

by Eduarda Oliosi, Afonso Caetano Júlio, Luís Silva, Phillip Probst, João Paulo Vilas-Boas, Ana Rita Pinheiro and Hugo Gamboa

Sensors 2025, 25(5), 1583; https://doi.org/10.3390/s25051583 - 5 Mar 2025

Viewed by 680

Abstract

This pilot study examines the relationship between pain intensity and trunk sitting postural control in 10 office workers with chronic spinal pain, using field-based real-time inertial sensors. Pain intensity was assessed with the Numeric Pain Rating Scale (NPRS) before and after work across [...] Read more.

This pilot study examines the relationship between pain intensity and trunk sitting postural control in 10 office workers with chronic spinal pain, using field-based real-time inertial sensors. Pain intensity was assessed with the Numeric Pain Rating Scale (NPRS) before and after work across three non-consecutive workdays, while postural control was evaluated through estimated center of pressure (COP) displacements. Linear and nonlinear metrics, including sway range, velocity, the Hurst exponent, and sample entropy, were derived from the estimated COP time series. Pearson correlation coefficients (r) and corresponding p-values were used to analyze the relationship between pain intensity and postural control. Significant correlations, though limited to specific metrics, were found (r = −0.860 to 0.855; p < 0.05), suggesting that higher pain intensity may be correlated with reduced postural variability. These findings provide preliminary insights into the potential link between pain intensity and postural control. Understanding trunk posture dynamics could inform the development of targeted ergonomic interventions to reduce musculoskeletal stress and improve sitting comfort in office environments. Full article

(This article belongs to the Special Issue Sensors for Physiological Monitoring and Digital Health)

► Show Figures

Figure 1

33 pages, 9852 KiB

Open AccessArticle

Assessment of Physiological Signals from Photoplethysmography Sensors Compared to an Electrocardiogram Sensor: A Validation Study in Daily Life

by Rana Zia Ur Rehman, Meenakshi Chatterjee, Nikolay V. Manyakov, Melina Daans, Amanda Jackson, Andrea O’Brisky, Tacie Telesky, Sophie Smets, Pieter-Jan Berghmans, Dongyan Yang, Elena Reynoso, Molly V. Lucas, Yanran Huo, Vasanth T. Thirugnanam, Tommaso Mansi and Mark Morris

Sensors 2024, 24(21), 6826; https://doi.org/10.3390/s24216826 - 24 Oct 2024

Cited by 2 | Viewed by 3117

Abstract

Wearables with photoplethysmography (PPG) sensors are being increasingly used in clinical research as a non-invasive, inexpensive method for remote monitoring of physiological health. Ensuring the accuracy and reliability of PPG-derived measurements is critical, as inaccuracies can impact research findings and clinical decisions. This [...] Read more.

Wearables with photoplethysmography (PPG) sensors are being increasingly used in clinical research as a non-invasive, inexpensive method for remote monitoring of physiological health. Ensuring the accuracy and reliability of PPG-derived measurements is critical, as inaccuracies can impact research findings and clinical decisions. This paper systematically compares heart rate (HR) and heart rate variability (HRV) measures from PPG against an electrocardiogram (ECG) monitor in free-living settings. Two devices with PPG and one device with an ECG sensor were worn by 25 healthy volunteers for 10 days. PPG-derived HR and HRV showed reasonable accuracy and reliability, particularly during sleep, with mean absolute error < 1 beat for HR and 6–15 ms for HRV. The relative error of HRV estimated from PPG varied with activity type and was higher than during the resting state by 14–51%. The accuracy of HR/HRV was impacted by the proportion of usable data, body posture, and epoch length. The multi-scale peak and trough detection algorithm demonstrated superior performance in detecting beats from PPG signals, with an F1 score of 89% during sleep. The study demonstrates the trade-offs of utilizing PPG measurements for remote monitoring in daily life and identifies optimal use conditions by recommending enhancements. Full article

(This article belongs to the Special Issue Sensors for Physiological Monitoring and Digital Health)

► Show Figures

Figure 1

12 pages, 2883 KiB

Open AccessArticle

Hybrid Integrated Wearable Patch for Brain EEG-fNIRS Monitoring

by Boyu Li, Mingjie Li, Jie Xia, Hao Jin, Shurong Dong and Jikui Luo

Sensors 2024, 24(15), 4847; https://doi.org/10.3390/s24154847 - 25 Jul 2024

Cited by 1 | Viewed by 2208

Abstract

Synchronous monitoring electroencephalogram (EEG) and functional near-infrared spectroscopy (fNIRS) have received significant attention in brain science research for their provision of more information on neuro-loop interactions. There is a need for an integrated hybrid EEG-fNIRS patch to synchronously monitor surface EEG and deep [...] Read more.

Synchronous monitoring electroencephalogram (EEG) and functional near-infrared spectroscopy (fNIRS) have received significant attention in brain science research for their provision of more information on neuro-loop interactions. There is a need for an integrated hybrid EEG-fNIRS patch to synchronously monitor surface EEG and deep brain fNIRS signals. Here, we developed a hybrid EEG-fNIRS patch capable of acquiring high-quality, co-located EEG and fNIRS signals. This patch is wearable and provides easy cognition and emotion detection, while reducing the spatial interference and signal crosstalk by integration, which leads to high spatial–temporal correspondence and signal quality. The modular design of the EEG-fNIRS acquisition unit and optimized mechanical design enables the patch to obtain EEG and fNIRS signals at the same location and eliminates spatial interference. The EEG pre-amplifier on the electrode side effectively improves the acquisition of weak EEG signals and significantly reduces input noise to 0.9 μV_rms, amplitude distortion to less than 2%, and frequency distortion to less than 1%. Detrending, motion correction algorithms, and band-pass filtering were used to remove physiological noise, baseline drift, and motion artifacts from the fNIRS signal. A high fNIRS source switching frequency configuration above 100 Hz improves crosstalk suppression between fNIRS and EEG signals. The Stroop task was carried out to verify its performance; the patch can acquire event-related potentials and hemodynamic information associated with cognition in the prefrontal area. Full article

(This article belongs to the Special Issue Sensors for Physiological Monitoring and Digital Health)

► Show Figures

Figure 1

19 pages, 3347 KiB

Open AccessArticle

Non-Invasive Blood Pressure Sensing via Machine Learning

by Filippo Attivissimo, Vito Ivano D’Alessandro, Luisa De Palma, Anna Maria Lucia Lanzolla and Attilio Di Nisio

Sensors 2023, 23(19), 8342; https://doi.org/10.3390/s23198342 - 9 Oct 2023

Cited by 13 | Viewed by 3760

Abstract

In this paper, a machine learning (ML) approach to estimate blood pressure (BP) using photoplethysmography (PPG) is presented. The final aim of this paper was to develop ML methods for estimating blood pressure (BP) in a non-invasive way that is suitable in a [...] Read more.

In this paper, a machine learning (ML) approach to estimate blood pressure (BP) using photoplethysmography (PPG) is presented. The final aim of this paper was to develop ML methods for estimating blood pressure (BP) in a non-invasive way that is suitable in a telemedicine health-care monitoring context. The training of regression models useful for estimating systolic blood pressure (SBP) and diastolic blood pressure (DBP) was conducted using new extracted features from PPG signals processed using the Maximal Overlap Discrete Wavelet Transform (MODWT). As a matter of fact, the interest was on the use of the most significant features obtained by the Minimum Redundancy Maximum Relevance (MRMR) selection algorithm to train eXtreme Gradient Boost (XGBoost) and Neural Network (NN) models. This aim was satisfactorily achieved by also comparing it with works in the literature; in fact, it was found that XGBoost models are more accurate than NN models in both systolic and diastolic blood pressure measurements, obtaining a Root Mean Square Error (RMSE) for SBP and DBP, respectively, of 5.67 mmHg and 3.95 mmHg. For SBP measurement, this result is an improvement compared to that reported in the literature. Furthermore, the trained XGBoost regression model fulfills the requirements of the Association for the Advancement of Medical Instrumentation (AAMI) as well as grade A of the British Hypertension Society (BHS) standard. Full article

(This article belongs to the Special Issue Sensors for Physiological Monitoring and Digital Health)

► Show Figures

Figure 1

17 pages, 5115 KiB

Open AccessArticle

Sensor Selection for Tidal Volume Determination via Linear Regression—Impact of Lasso versus Ridge Regression

by Bernhard Laufer, Paul D. Docherty, Rua Murray, Sabine Krueger-Ziolek, Nour Aldeen Jalal, Fabian Hoeflinger, Stefan J. Rupitsch, Leonhard Reindl and Knut Moeller

Sensors 2023, 23(17), 7407; https://doi.org/10.3390/s23177407 - 25 Aug 2023

Cited by 5 | Viewed by 1713

Abstract

The measurement of respiratory volume based on upper body movements by means of a smart shirt is increasingly requested in medical applications. This research used upper body surface motions obtained by a motion capture system, and two regression methods to determine the optimal [...] Read more.

The measurement of respiratory volume based on upper body movements by means of a smart shirt is increasingly requested in medical applications. This research used upper body surface motions obtained by a motion capture system, and two regression methods to determine the optimal selection and placement of sensors on a smart shirt to recover respiratory parameters from benchmark spirometry values. The results of the two regression methods (Ridge regression and the least absolute shrinkage and selection operator (Lasso)) were compared. This work shows that the Lasso method offers advantages compared to the Ridge regression, as it provides sparse solutions and is more robust to outliers. However, both methods can be used in this application since they lead to a similar sensor subset with lower computational demand (from exponential effort for full exhaustive search down to the order of O (n²)). A smart shirt for respiratory volume estimation could replace spirometry in some cases and would allow for a more convenient measurement of respiratory parameters in home care or hospital settings. Full article

(This article belongs to the Special Issue Sensors for Physiological Monitoring and Digital Health)

► Show Figures

Figure 1

19 pages, 11434 KiB

Open AccessArticle

Simultaneous Sleep Stage and Sleep Disorder Detection from Multimodal Sensors Using Deep Learning

by Yi-Hsuan Cheng, Margaret Lech and Richardt Howard Wilkinson

Sensors 2023, 23(7), 3468; https://doi.org/10.3390/s23073468 - 26 Mar 2023

Cited by 15 | Viewed by 4085

Abstract

Sleep scoring involves the inspection of multimodal recordings of sleep data to detect potential sleep disorders. Given that symptoms of sleep disorders may be correlated with specific sleep stages, the diagnosis is typically supported by the simultaneous identification of a sleep stage and [...] Read more.

Sleep scoring involves the inspection of multimodal recordings of sleep data to detect potential sleep disorders. Given that symptoms of sleep disorders may be correlated with specific sleep stages, the diagnosis is typically supported by the simultaneous identification of a sleep stage and a sleep disorder. This paper investigates the automatic recognition of sleep stages and disorders from multimodal sensory data (EEG, ECG, and EMG). We propose a new distributed multimodal and multilabel decision-making system (MML-DMS). It comprises several interconnected classifier modules, including deep convolutional neural networks (CNNs) and shallow perceptron neural networks (NNs). Each module works with a different data modality and data label. The flow of information between the MML-DMS modules provides the final identification of the sleep stage and sleep disorder. We show that the fused multilabel and multimodal method improves the diagnostic performance compared to single-label and single-modality approaches. We tested the proposed MML-DMS on the PhysioNet CAP Sleep Database, with VGG16 CNN structures, achieving an average classification accuracy of 94.34% and

F_{1}

score of 0.92 for sleep stage detection (six stages) and an average classification accuracy of 99.09% and

F_{1}

score of 0.99 for sleep disorder detection (eight disorders). A comparison with related studies indicates that the proposed approach significantly improves upon the existing state-of-the-art approaches. Full article

(This article belongs to the Special Issue Sensors for Physiological Monitoring and Digital Health)

► Show Figures

Figure 1

21 pages, 868 KiB

Open AccessArticle

SenseHunger: Machine Learning Approach to Hunger Detection Using Wearable Sensors

by Muhammad Tausif Irshad, Muhammad Adeel Nisar, Xinyu Huang, Jana Hartz, Olaf Flak, Frédéric Li, Philip Gouverneur, Artur Piet, Kerstin M. Oltmanns and Marcin Grzegorzek

Sensors 2022, 22(20), 7711; https://doi.org/10.3390/s22207711 - 11 Oct 2022

Cited by 17 | Viewed by 5866

Abstract

The perception of hunger and satiety is of great importance to maintaining a healthy body weight and avoiding chronic diseases such as obesity, underweight, or deficiency syndromes due to malnutrition. There are a number of disease patterns, characterized by a chronic loss of [...] Read more.

The perception of hunger and satiety is of great importance to maintaining a healthy body weight and avoiding chronic diseases such as obesity, underweight, or deficiency syndromes due to malnutrition. There are a number of disease patterns, characterized by a chronic loss of this perception. To our best knowledge, hunger and satiety cannot be classified using non-invasive measurements. Aiming to develop an objective classification system, this paper presents a multimodal sensory system using associated signal processing and pattern recognition methods for hunger and satiety detection based on non-invasive monitoring. We used an Empatica E4 smartwatch, a RespiBan wearable device, and JINS MEME smart glasses to capture physiological signals from five healthy normal weight subjects inactively sitting on a chair in a state of hunger and satiety. After pre-processing the signals, we compared different feature extraction approaches, either based on manual feature engineering or deep feature learning. Comparative experiments were carried out to determine the most appropriate sensor channel, device, and classifier to reliably discriminate between hunger and satiety states. Our experiments showed that the most discriminative features come from three specific sensor modalities: Electrodermal Activity (EDA), infrared Thermopile (Tmp), and Blood Volume Pulse (BVP). Full article

(This article belongs to the Special Issue Sensors for Physiological Monitoring and Digital Health)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Sensors for Physiological Monitoring and Digital Health

Share This Special Issue

Special Issue Editors

Special Issue Information

Benefits of Publishing in a Special Issue

Published Papers (8 papers)

Research

Further Information

Guidelines

MDPI Initiatives

Follow MDPI