Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
5.3
2.6
Journals
Electronics
Special Issues
Machine Learning for Biomedical Applications
share announcement

Machine Learning for Biomedical Applications

A special issue of Electronics (ISSN 2079-9292). This special issue belongs to the section "Bioelectronics".

Deadline for manuscript submissions: closed (20 November 2024) | Viewed by 12228

Special Issue Editor

Dr. William M. Mongan

E-Mail Website
Guest Editor
Department of Mathematics and Computer Science, Ursinus College, Collegeville, PA 19426, USA
Interests: machine learning; passive RF wearable systems for biomedical classification; IoT sensor-actuator systems; computer science education

Special Issue Information

Dear Colleagues,

This Special Issue of Electronics on Machine Learning for Biomedical Applications aims to showcase and synthesize the pillars of the end-to-end connected pipeline of sensor data collection and/or storage, processing, decision-making, actuation, and security of biomedical sensor systems. In this issue, we will explore current trends and challenges in biomedical systems and machine learning tools and techniques for processing their data in biomedically meaningful ways, with a specific aim of facilitating practical application, expansion, and scale of the techniques presented. The purpose of this Special Issue is to enable researchers to a) rapidly define the current state-of-the-art in wireless/ubiquitous sensor systems for biomedical applications, b) to apply machine learning to wireless/ubiquitous sensor signals for real-time or near-real-time decisioning, and c) to identify current trends and gaps for future work in learning techniques, security and privacy, cloud- and edge processing, and considerations for efficient computing in constrained environments. An omnipresent concern underlying each of these facets is the need for responsible machine learning and transparent decisioning as well as effective human–machine interfaces to promote a safe and seamless user experience, and this issue aims to highlight the current state and ongoing need for responsible and ethical development and use of machine learning. Accepted articles will build upon existing literature by offering surveys of existing work across each pillar of the ecosystem and showcasing compelling biomedical applications that build upon leading-edge tools and techniques as well as the unique challenges of applying machine learning to biomedical applications such as generalization of training data, using unsupervised learning for real-time and near-real-time applications, and adapting estimates to noisy ground truth observations. In addition, wireless systems present unique challenges including security and privacy considerations on collected data, power constraints on ubiquitous, wearable, or passive wireless sensors, and shifting computation between the cloud, edge, and hybrid environments as well as between the physical and processing layers. By exploring novel solutions to these unique challenges, we seek to enhance the current state of machine learning systems in biomedical applications as well as basic research in machine learning techniques broadly. Thus, this Special Issue presents a holistic view of human-centric machine learning on biomedical systems.

Dr. William M. Mongan 
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Electronics is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • biomedical sensor systems
  • biomedical machine learning
  • wireless Internet-of-Things sensors and actuators
  • wearable sensors
  • wireless biomedical security and privacy
  • passive and energy-constrained biomedical sensor systems
  • human-centric biomedical wireless machine learning

Benefits of Publishing in a Special Issue

  • Ease of navigation: Grouping papers by topic helps scholars navigate broad scope journals more efficiently.
  • Greater discoverability: Special Issues support the reach and impact of scientific research. Articles in Special Issues are more discoverable and cited more frequently.
  • Expansion of research network: Special Issues facilitate connections among authors, fostering scientific collaborations.
  • External promotion: Articles in Special Issues are often promoted through the journal's social media, increasing their visibility.
  • e-Book format: Special Issues with more than 10 articles can be published as dedicated e-books, ensuring wide and rapid dissemination.

Further information on MDPI's Special Issue policies can be found here.

Published Papers (7 papers)

Download All Papers
Order results
Result details
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:

Research

18 pages, 4318 KiB  
Article
Data-Driven Maturity Level Evaluation for Cardiomyocytes Derived from Human Pluripotent Stem Cells (Invited Paper)
by Yan Hong, Xueqing Huang, Fang Li, Siqi Huang, Qibiao Weng, Diego Fraidenraich and Ioana Voiculescu
Electronics 2024, 13(24), 4985; https://doi.org/10.3390/electronics13244985 - 18 Dec 2024
Viewed by 842
Abstract
Cardiovascular disease is a leading cause of death worldwide. The differentiation of human pluripotent stem cells (hPSCs) into functional cardiomyocytes offers significant potential for disease modeling and cell-based cardiac therapies. However, hPSC-derived cardiomyocytes (hPSC-CMs) remain largely immature, limiting their experimental and clinical applications. [...] Read more.
Cardiovascular disease is a leading cause of death worldwide. The differentiation of human pluripotent stem cells (hPSCs) into functional cardiomyocytes offers significant potential for disease modeling and cell-based cardiac therapies. However, hPSC-derived cardiomyocytes (hPSC-CMs) remain largely immature, limiting their experimental and clinical applications. A critical challenge in current in vitro culture systems is the absence of standardized metrics to quantify maturity. This study presents a data-driven pipeline to quantify hPSC-CM maturity using gene expression data across various stages of cardiac development. We determined that culture time serves as a feasible proxy for maturity. To improve prediction accuracy, machine learning algorithms were employed to identify heart-related genes whose expression strongly correlates with culture time. Our results reduced the average discrepancy between predicted and observed culture time to 4.461 days and CASQ2 (Calsequestrin 2), a gene involved in calcium ion storage and transport, was identified as the most critical cardiac gene associated with culture duration. This novel framework for maturity assessment moves beyond traditional qualitative methods, providing deeper insights into hPSC-CM maturation dynamics. It establishes a foundation for developing advanced lab-on-chip devices capable of real-time maturity monitoring and adaptive stimulus selection, paving the way for improved maturation strategies and broader experimental/clinical applications. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
►▼ Show Figures

Figure 1

19 pages, 3414 KiB  
Article
Deep Contrastive Survival Analysis with Dual-View Clustering
by Chang Cui, Yongqiang Tang and Wensheng Zhang
Electronics 2024, 13(24), 4866; https://doi.org/10.3390/electronics13244866 - 10 Dec 2024
Cited by 1 | Viewed by 790
Abstract
Survival analysis aims to analyze the relationship between covariates and events of interest, and is widely applied in multiple research fields, especially in clinical fields. Recently, some studies have attempted to discover potential sub-populations in survival data to assist in survival prediction with [...] Read more.
Survival analysis aims to analyze the relationship between covariates and events of interest, and is widely applied in multiple research fields, especially in clinical fields. Recently, some studies have attempted to discover potential sub-populations in survival data to assist in survival prediction with clustering. However, existing models that combine clustering with survival analysis face multiple challenges: incomplete representation caused by single-path encoders, the incomplete information of pseudo-samples, and misleading effects of boundary samples. To overcome these challenges, in this study, we propose a novel deep contrastive survival analysis model with dual-view clustering. Specifically, we design a Siamese autoencoder to construct latent spaces in two views and conduct dual-view clustering to more comprehensively capture patient representations. Moreover, we consider the dual views as mutual augmentations rather than introducing pseudo-samples and, based on this, triplet contrastive learning is proposed to fully utilize clustering information and dual-view representations to enhance survival prediction. Additionally, we employ a self-paced learning strategy in the dual-view clustering process to ensure the model handles samples from easy to hard in training, thereby avoiding the misleading effects of boundary samples. Our proposal achieves an average C-index and IBS of 0.6653 and 0.1786 on three widely used clinical datasets, both exceeding the existing best methods, which demonstrates its advanced discriminative and calibration performance. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
►▼ Show Figures

Figure 1

14 pages, 4683 KiB  
Article
Wearable System for Continuous Estimation of Transepidermal Water Loss
by Natsumi Matsui, Ayumi Ohnishi, Ayaka Uyama, Tetsuzo Sugino, Tsutomu Terada and Masahiko Tsukamoto
Electronics 2024, 13(23), 4779; https://doi.org/10.3390/electronics13234779 - 3 Dec 2024
Cited by 1 | Viewed by 1020
Abstract
To maintain skin moisture, we need to maintain good stratum corneum barrier function, which prevents moisture evaporation from the inside of the skin. Transepidermal water loss (TEWL), the amount of water that evaporates from the skin, indicates the state of barrier function. The [...] Read more.
To maintain skin moisture, we need to maintain good stratum corneum barrier function, which prevents moisture evaporation from the inside of the skin. Transepidermal water loss (TEWL), the amount of water that evaporates from the skin, indicates the state of barrier function. The barrier function of facial skin is easily damaged in daily life, and the condition of the skin becomes worse without us noticing. We should constantly monitor TEWL to prevent worsening skin conditions. In this paper, we propose a wearable device that continuously measures TEWL. We estimate TEWL using machine learning from temperature and humidity values of water evaporation from the skin and parameters that affect TEWL, such as skin surface temperature and galvanic skin response. We experimented with the prototype device in a controlled environment. We confirmed that the prototype device could estimate TEWL accurately enough to judge the skin’s condition in stationary and conversational situations. Then, we experimented to verify the environmental conditions for estimating TEWL using the prototype device. The prototype device could estimate TEWL with sufficient precision in an office without airflow. However, we could not estimate TEWL in the office with airflow and outdoor. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
►▼ Show Figures

Figure 1

18 pages, 3413 KiB  
Article
Diagnosing and Characterizing Chronic Kidney Disease with Machine Learning: The Value of Clinical Patient Characteristics as Evidenced from an Open Dataset
by Juan Figueroa, Patrick Etim, Adithyan Karanathu Shibu, Derek Berger and Jacob Levman
Electronics 2024, 13(21), 4326; https://doi.org/10.3390/electronics13214326 - 4 Nov 2024
Cited by 1 | Viewed by 2202
Abstract
Applying artificial intelligence (AI) and machine learning for chronic kidney disease (CKD) diagnostics and characterization has the potential to improve the standard of patient care through accurate and early detection, as well as providing a more detailed understanding of the condition. This study [...] Read more.
Applying artificial intelligence (AI) and machine learning for chronic kidney disease (CKD) diagnostics and characterization has the potential to improve the standard of patient care through accurate and early detection, as well as providing a more detailed understanding of the condition. This study employed reproducible validation of AI technology with public domain software applied to CKD diagnostics on a publicly available CKD dataset acquired from 400 patients. The approach presented includes patient-specific symptomatic variables and demonstrates performance improvements associated with this approach. Our best-performing AI models, which include patient symptom variables, achieve predictive accuracies ranging from 99.4 to 100% across both hold-out and 5-fold validation with the light gradient boosting machine. We demonstrate that the exclusion of patient symptom variables reduces model performance in line with the literature on the same dataset. We also provide an unsupervised learning cluster analysis to help interpret variability among, and characterize the population of, patients with CKD. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
►▼ Show Figures

Figure 1

21 pages, 6541 KiB  
Article
Comparison of Machine Learning Models for Predicting Interstitial Glucose Using Smart Watch and Food Log
by Haider Ali, Imran Khan Niazi, David White, Malik Naveed Akhter and Samaneh Madanian
Electronics 2024, 13(16), 3192; https://doi.org/10.3390/electronics13163192 - 12 Aug 2024
Cited by 1 | Viewed by 2205
Abstract
This study examines the performance of various machine learning (ML) models in predicting Interstitial Glucose (IG) levels using data from wrist-worn wearable sensors. The insights from these predictions can aid in understanding metabolic syndromes and disease states. A public dataset comprising information from [...] Read more.
This study examines the performance of various machine learning (ML) models in predicting Interstitial Glucose (IG) levels using data from wrist-worn wearable sensors. The insights from these predictions can aid in understanding metabolic syndromes and disease states. A public dataset comprising information from the Empatica E4 smart watch, the Dexcom Continuous Glucose Monitor (CGM) measuring IG, and a food log was utilized. The raw data were processed into features, which were then used to train different ML models. This study evaluates the performance of decision tree (DT), support vector machine (SVM), Random Forest (RF), Linear Discriminant Analysis (LDA), K-Nearest Neighbors (KNN), Gaussian Naïve Bayes (GNB), lasso cross-validation (LassoCV), Ridge, Elastic Net, and XGBoost models. For classification, IG labels were categorized into high, standard, and low, and the performance of the ML models was assessed using accuracy (40–78%), precision (41–78%), recall (39–77%), F1-score (0.31–0.77), and receiver operating characteristic (ROC) curves. Regression models predicting IG values were evaluated based on R-squared values (−7.84–0.84), mean absolute error (5.54–60.84 mg/dL), root mean square error (9.04–68.07 mg/dL), and visual methods like residual and QQ plots. To assess whether the differences between models were statistically significant, the Friedman test was carried out and was interpreted using the Nemenyi post hoc test. Tree-based models, particularly RF and DT, demonstrated superior accuracy for classification tasks in comparison to other models. For regression, the RF model achieved the lowest RMSE of 9.04 mg/dL with an R-squared value of 0.84, while the GNB model performed the worst, with an RMSE of 68.07 mg/dL. A SHAP analysis identified time from midnight as the most significant predictor. Partial dependence plots revealed complex feature interactions in the RF model, contrasting with the simpler interactions captured by LDA. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
►▼ Show Figures

Graphical abstract

18 pages, 2964 KiB  
Article
Initial Development and Analysis of a Context-Aware Burn Resuscitation Decision-Support Algorithm
by Yi-Ming Kao, Ghazal Arabidarrehdor, Babita Parajuli, Eriks E. Ziedins, Melissa M. McLawhorn, Cameron S. D’Orio, Mary Oliver, Lauren Moffatt, Shane K. Mathew, Edward J. Kelly, Bonnie C. Carney, Jeffrey W. Shupp, David M. Burmeister and Jin-Oh Hahn
Electronics 2024, 13(14), 2713; https://doi.org/10.3390/electronics13142713 - 11 Jul 2024
Cited by 1 | Viewed by 1069
Abstract
Burn patients require high-volume intravenous resuscitation with the goal of restoring global tissue perfusion to make up for burn-induced loss of fluid from the vasculature. Clinical standards of burn resuscitation are predominantly based on urinary output, which is not context-aware because it is [...] Read more.
Burn patients require high-volume intravenous resuscitation with the goal of restoring global tissue perfusion to make up for burn-induced loss of fluid from the vasculature. Clinical standards of burn resuscitation are predominantly based on urinary output, which is not context-aware because it is not a trustworthy indicator of tissue perfusion. This paper investigates the initial development and analysis of a context-aware decision-support algorithm for burn resuscitation. In this context, we hypothesized that the use of a more context-aware surrogate of tissue perfusion may enhance the efficacy of burn resuscitation in normalizing cardiac output. Toward this goal, we exploited the arterial pulse wave analysis to discover novel surrogates of cardiac output. Then, we developed the cardiac output-enabled burn resuscitation decision-support (CaRD) algorithm. Using experimental data collected from animals undergoing burn injury and resuscitation, we conducted an initial evaluation and analysis of the CaRD algorithm in comparison with the commercially available Burn NavigatorTM algorithm. Combining a surrogate of cardiac output with urinary output in the CaRD algorithm has the potential to improve the efficacy of burn resuscitation. However, the improvement achieved in this work was only marginal, which is likely due to the suboptimal tuning of the CaRD algorithm with the limited available dataset. In this way, the results showed both promise and challenges that are crucial to future algorithm development. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
►▼ Show Figures

Figure 1

21 pages, 2543 KiB  
Article
Assessing the Reliability of Machine Learning Models Applied to the Mental Health Domain Using Explainable AI
by Vishnu Pendyala and Hyungkyun Kim
Electronics 2024, 13(6), 1025; https://doi.org/10.3390/electronics13061025 - 8 Mar 2024
Cited by 4 | Viewed by 3022
Abstract
Machine learning is increasingly and ubiquitously being used in the medical domain. Evaluation metrics like accuracy, precision, and recall may indicate the performance of the models but not necessarily the reliability of their outcomes. This paper assesses the effectiveness of a number of [...] Read more.
Machine learning is increasingly and ubiquitously being used in the medical domain. Evaluation metrics like accuracy, precision, and recall may indicate the performance of the models but not necessarily the reliability of their outcomes. This paper assesses the effectiveness of a number of machine learning algorithms applied to an important dataset in the medical domain, specifically, mental health, by employing explainability methodologies. Using multiple machine learning algorithms and model explainability techniques, this work provides insights into the models’ workings to help determine the reliability of the machine learning algorithm predictions. The results are not intuitive. It was found that the models were focusing significantly on less relevant features and, at times, unsound ranking of the features to make the predictions. This paper therefore argues that it is important for research in applied machine learning to provide insights into the explainability of models in addition to other performance metrics like accuracy. This is particularly important for applications in critical domains such as healthcare. Full article
(This article belongs to the Special Issue Machine Learning for Biomedical Applications)
►▼ Show Figures

Figure 1

Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-7
Back to TopTop