Submit to Bioengineering Review for Bioengineering Propose a Special Issue

Journal Menu

Journal Browser

► Journal Browser

Data Processing and Machine Learning for Assistive and Rehabilitation Technologies

Special Issue Editors
Special Issue Information
Keywords
Benefits of Publishing in a Special Issue
Published Papers

A special issue of Bioengineering (ISSN 2306-5354). This special issue belongs to the section "Biosignal Processing".

Deadline for manuscript submissions: closed (30 April 2023) | Viewed by 30510

Share This Special Issue

Special Issue Editors

Dr. Agnese Sbrollini

E-Mail Website
Guest Editor

Department of Information Engineering, Università Politecnica delle Marche, Via Brecce Bianche 12, 60131 Ancona, Italy
Interests: cardiac signal processing; biostatistics applied to cardiac signals; artificial intelligence in medicine and biology; clinical decision support systems; wearable and portable sensors; cardiorespiratory monitoring in sport; serial electrocardiography; atrial fibrillation; fetal and newborn monitoring
Special Issues, Collections and Topics in MDPI journals

Dr. Alessandro Mengarelli

E-Mail Website
Guest Editor

Deparment of Information Engineering, Università Politecnica delle Marche, Via Brecce Bianche 12, 60131 Ancona, Italy
Interests: motion analysis; movement biomechanics; biosignal processing and modeling; control of biomechanical systems
Special Issues, Collections and Topics in MDPI journals

Dr. Andrea Tigrini

E-Mail Website
Guest Editor

Deparment of Information Engineering, Università Politecnica delle Marche, Via Brecce Bianche 12, 60131 Ancona, Italy
Interests: human biomechanics; neuromuscular control and rehabilitation engineering
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

In the past few years, the rehabilitation field underwent a dramatic advancement guided by the availability of new sensing technologies and smart embedded computing (i.e., wearable devices), allowing the acquisition and processing of a wide range of physiological data. This innovation supported the development of architectures used to assist clinicians in the diagnosis and treatment of pathologies that involve the neurological, musculoskeletal, and cardiorespiratory systems. This has provided new opportunities for the design and development of subject-oriented rehabilitative pathways characterized by quantitative analyses rather than clinical experience-based assessments. In this context, the literature highlights an increasing demand of technical solutions that allow the processing of data and decisions quickly, guaranteeing real time feedback information for the users and to the assistive architectures. Thus, the realization of algorithms for data processing and machine learning characterized by low computational costs represents a sizeable challenge in this field. This supports the research line for the development of new filtering techniques that fuse multiple sensors’ information and extract new patterns to support clinicians in rehabilitation. This information can also be used to identify data-driven models of the detrimental effects due to a given pathology, eventually predicting the patient health state to prevent critical events. Hence, the aim of this Special Issue is to collect original research and review papers regarding the use of data processing and machine learning techniques that allow us to overcome the limits of the already available rehabilitative treatments. Submissions can also include the development of filtering techniques and machine learning models with a particular focus on real time applications for assistive technologies. Moreover, research involving data-driven models for the treatment of neurological, musculoskeletal, and cardiorespiratory disorders or disease are particularly welcome.

Potential topics include but are not limited to the following:

real time data processing in rehabilitation scenarios;
machine learning and data-driven models for treatment delivery;
artificial intelligence in medical robotics;
neurological, musculoskeletal, and cardiorespiratory diseases effects modeling;
architectures for telemedicine;
computer-aided rehabilitation;
smart assistive technologies;
data patterns discovery for improving rehabilitation treatments;
pattern recognition based neuromuscular control;
embedded machine learning for medical technologies.

Dr. Agnese Sbrollini
Dr. Alessandro Mengarelli
Dr. Andrea Tigrini
Guest Editors

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. Bioengineering is an international peer-reviewed open access monthly 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 2700 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

rehabilitation
telemedicine
robotics
biomedical signal processing
embedded machine learning
neuromuscular control
assistive technologies

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 polices can be found here.

Published Papers (9 papers)

Download All Papers

Order results

Result details

Show export options Show export options

Select all

Export citation of selected articles as:

Research

Jump to: Other

22 pages, 7274 KiB

Open AccessArticle

An Approach to Binary Classification of Alzheimer’s Disease Using LSTM

by Waleed Salehi, Preety Baglat, Gaurav Gupta, Surbhi Bhatia Khan, Ahlam Almusharraf, Ali Alqahtani and Adarsh Kumar

Bioengineering 2023, 10(8), 950; https://doi.org/10.3390/bioengineering10080950 - 9 Aug 2023

Cited by 5 | Viewed by 2484

Abstract

In this study, we use LSTM (Long-Short-Term-Memory) networks to evaluate Magnetic Resonance Imaging (MRI) data to overcome the shortcomings of conventional Alzheimer’s disease (AD) detection techniques. Our method offers greater reliability and accuracy in predicting the possibility of AD, in contrast to cognitive testing and brain structure analyses. We used an MRI dataset that we downloaded from the Kaggle source to train our LSTM network. Utilizing the temporal memory characteristics of LSTMs, the network was created to efficiently capture and evaluate the sequential patterns inherent in MRI scans. Our model scored a remarkable AUC of 0.97 and an accuracy of 98.62%. During the training process, we used Stratified Shuffle-Split Cross Validation to make sure that our findings were reliable and generalizable. Our study adds significantly to the body of knowledge by demonstrating the potential of LSTM networks in the specific field of AD prediction and extending the variety of methods investigated for image classification in AD research. We have also designed a user-friendly Web-based application to help with the accessibility of our developed model, bridging the gap between research and actual deployment. Full article

(This article belongs to the Special Issue Data Processing and Machine Learning for Assistive and Rehabilitation Technologies)

► Show Figures

Figure 1

17 pages, 3227 KiB

Open AccessArticle

Data-Driven Quantitation of Movement Abnormality after Stroke

by Avinash Parnandi, Aakash Kaku, Anita Venkatesan, Natasha Pandit, Emily Fokas, Boyang Yu, Grace Kim, Dawn Nilsen, Carlos Fernandez-Granda and Heidi Schambra

Bioengineering 2023, 10(6), 648; https://doi.org/10.3390/bioengineering10060648 - 26 May 2023

Cited by 1 | Viewed by 1913

Abstract

Stroke commonly affects the ability of the upper extremities (UEs) to move normally. In clinical settings, identifying and measuring movement abnormality is challenging due to the imprecision and impracticality of available assessments. These challenges interfere with therapeutic tracking, communication, and treatment. We thus sought to develop an approach that blends precision and pragmatism, combining high-dimensional motion capture with out-of-distribution (OOD) detection. We used an array of wearable inertial measurement units to capture upper body motion in healthy and chronic stroke subjects performing a semi-structured, unconstrained 3D tabletop task. After data were labeled by human coders, we trained two deep learning models exclusively on healthy subject data to classify elemental movements (functional primitives). We tested these healthy subject-trained models on previously unseen healthy and stroke motion data. We found that model confidence, indexed by prediction probabilities, was generally high for healthy test data but significantly dropped when encountering OOD stroke data. Prediction probabilities worsened with more severe motor impairment categories and were directly correlated with individual impairment scores. Data inputs from the paretic UE, rather than trunk, most strongly influenced model confidence. We demonstrate for the first time that using OOD detection with high-dimensional motion data can reveal clinically meaningful movement abnormality in subjects with chronic stroke. Full article

(This article belongs to the Special Issue Data Processing and Machine Learning for Assistive and Rehabilitation Technologies)

► Show Figures

Graphical abstract

18 pages, 8060 KiB

Open AccessArticle

Empowering Hand Rehabilitation with AI-Powered Gesture Recognition: A Study of an sEMG-Based System

by Kai Guo, Mostafa Orban, Jingxin Lu, Maged S. Al-Quraishi, Hongbo Yang and Mahmoud Elsamanty

Bioengineering 2023, 10(5), 557; https://doi.org/10.3390/bioengineering10050557 - 6 May 2023

Cited by 7 | Viewed by 4883

Abstract

Stroke is one of the most prevalent health issues that people face today, causing long-term complications such as paresis, hemiparesis, and aphasia. These conditions significantly impact a patient’s physical abilities and cause financial and social hardships. In order to address these challenges, this paper presents a groundbreaking solution—a wearable rehabilitation glove. This motorized glove is designed to provide comfortable and effective rehabilitation for patients with paresis. Its unique soft materials and compact size make it easy to use in clinical settings and at home. The glove can train each finger individually and all fingers together, using assistive force generated by advanced linear integrated actuators controlled by sEMG signals. The glove is also durable and long-lasting, with 4–5 h of battery life. The wearable motorized glove is worn on the affected hand to provide assistive force during rehabilitation training. The key to this glove’s effectiveness is its ability to perform the classified hand gestures acquired from the non-affected hand by integrating four sEMG sensors and a deep learning algorithm (the 1D-CNN algorithm and the InceptionTime algorithm). The InceptionTime algorithm classified ten hand gestures’ sEMG signals with an accuracy of 91.60% and 90.09% in the training and verification sets, respectively. The overall accuracy was 90.89%. It showed potential as a tool for developing effective hand gesture recognition systems. The classified hand gestures can be used as a control command for the motorized wearable glove placed on the affected hand, allowing it to mimic the movements of the non-affected hand. This innovative technology performs rehabilitation exercises based on the theory of mirror therapy and task-oriented therapy. Overall, this wearable rehabilitation glove represents a significant step forward in stroke rehabilitation, offering a practical and effective solution to help patients recover from stroke’s physical, financial, and social impact. Full article

(This article belongs to the Special Issue Data Processing and Machine Learning for Assistive and Rehabilitation Technologies)

► Show Figures

Figure 1

14 pages, 2217 KiB

Open AccessArticle

Fast Optical Signals for Real-Time Retinotopy and Brain Computer Interface

by David Perpetuini, Mehmet Günal, Nicole Chiou, Sanmi Koyejo, Kyle Mathewson, Kathy A. Low, Monica Fabiani, Gabriele Gratton and Antonio Maria Chiarelli

Bioengineering 2023, 10(5), 553; https://doi.org/10.3390/bioengineering10050553 - 5 May 2023

Cited by 3 | Viewed by 2082

Abstract

A brain–computer interface (BCI) allows users to control external devices through brain activity. Portable neuroimaging techniques, such as near-infrared (NIR) imaging, are suitable for this goal. NIR imaging has been used to measure rapid changes in brain optical properties associated with neuronal activation, namely fast optical signals (FOS) with good spatiotemporal resolution. However, FOS have a low signal-to-noise ratio, limiting their BCI application. Here FOS were acquired with a frequency-domain optical system from the visual cortex during visual stimulation consisting of a rotating checkerboard wedge, flickering at 5 Hz. We used measures of photon count (Direct Current, DC light intensity) and time of flight (phase) at two NIR wavelengths (690 nm and 830 nm) combined with a machine learning approach for fast estimation of visual-field quadrant stimulation. The input features of a cross-validated support vector machine classifier were computed as the average modulus of the wavelet coherence between each channel and the average response among all channels in 512 ms time windows. An above chance performance was obtained when differentiating visual stimulation quadrants (left vs. right or top vs. bottom) with the best classification accuracy of ~63% (information transfer rate of ~6 bits/min) when classifying the superior and inferior stimulation quadrants using DC at 830 nm. The method is the first attempt to provide generalizable retinotopy classification relying on FOS, paving the way for the use of FOS in real-time BCI. Full article

(This article belongs to the Special Issue Data Processing and Machine Learning for Assistive and Rehabilitation Technologies)

► Show Figures

Figure 1

16 pages, 2185 KiB

Open AccessArticle

Machine Learning and Explainable Artificial Intelligence Using Counterfactual Explanations for Evaluating Posture Parameters

by Carlo Dindorf, Oliver Ludwig, Steven Simon, Stephan Becker and Michael Fröhlich

Bioengineering 2023, 10(5), 511; https://doi.org/10.3390/bioengineering10050511 - 24 Apr 2023

Cited by 5 | Viewed by 2088

Abstract

Postural deficits such as hyperlordosis (hollow back) or hyperkyphosis (hunchback) are relevant health issues. Diagnoses depend on the experience of the examiner and are, therefore, often subjective and prone to errors. Machine learning (ML) methods in combination with explainable artificial intelligence (XAI) tools have proven useful for providing an objective, data-based orientation. However, only a few works have considered posture parameters, leaving the potential for more human-friendly XAI interpretations still untouched. Therefore, the present work proposes an objective, data-driven ML system for medical decision support that enables especially human-friendly interpretations using counterfactual explanations (CFs). The posture data for 1151 subjects were recorded by means of stereophotogrammetry. An expert-based classification of the subjects regarding the presence of hyperlordosis or hyperkyphosis was initially performed. Using a Gaussian progress classifier, the models were trained and interpreted using CFs. The label errors were flagged and re-evaluated using confident learning. Very good classification performances for both hyperlordosis and hyperkyphosis were found, whereby the re-evaluation and correction of the test labels led to a significant improvement (M_PRAUC = 0.97). A statistical evaluation showed that the CFs seemed to be plausible, in general. In the context of personalized medicine, the present study’s approach could be of importance for reducing diagnostic errors and thereby improving the individual adaptation of therapeutic measures. Likewise, it could be a basis for the development of apps for preventive posture assessment. Full article

(This article belongs to the Special Issue Data Processing and Machine Learning for Assistive and Rehabilitation Technologies)

► Show Figures

Graphical abstract

18 pages, 3863 KiB

Open AccessEditor’s ChoiceArticle

Exploring the Role of Visual Guidance in Motor Imagery-Based Brain-Computer Interface: An EEG Microstate-Specific Functional Connectivity Study

by Tianjun Wang, Yun-Hsuan Chen and Mohamad Sawan

Bioengineering 2023, 10(3), 281; https://doi.org/10.3390/bioengineering10030281 - 21 Feb 2023

Cited by 5 | Viewed by 2942

Abstract

Motor imagery-based brain–computer interfaces (BCI) have been widely recognized as beneficial tools for rehabilitation applications. Moreover, visually guided motor imagery was introduced to improve the rehabilitation impact. However, the reported results to support these techniques remain unsatisfactory. Electroencephalography (EEG) signals can be represented by a sequence of a limited number of topographies (microstates). To explore the dynamic brain activation patterns, we conducted EEG microstate and microstate-specific functional connectivity analyses on EEG data under motor imagery (MI), motor execution (ME), and guided MI (GMI) conditions. By comparing sixteen microstate parameters, the brain activation patterns induced by GMI show more similarities to ME than MI from a microstate perspective. The mean duration and duration of microstate four are proposed as biomarkers to evaluate motor condition. A support vector machine (SVM) classifier trained with microstate parameters achieved average accuracies of 80.27% and 66.30% for ME versus MI and GMI classification, respectively. Further, functional connectivity patterns showed a strong relationship with microstates. Key node analysis shows clear switching of key node distribution between brain areas among different microstates. The neural mechanism of the switching pattern is discussed. While microstate analysis indicates similar brain dynamics between GMI and ME, graph theory-based microstate-specific functional connectivity analysis implies that visual guidance may reduce the functional integration of the brain network during MI. Thus, we proposed that combined MI and GMI for BCI can improve neurorehabilitation effects. The present findings provide insights for understanding the neural mechanism of microstates, the role of visual guidance in MI tasks, and the experimental basis for developing new BCI-aided rehabilitation systems. Full article

(This article belongs to the Special Issue Data Processing and Machine Learning for Assistive and Rehabilitation Technologies)

► Show Figures

Graphical abstract

22 pages, 6159 KiB

Open AccessArticle

A New Wrist–Forearm Rehabilitation Protocol Integrating Human Biomechanics and SVM-Based Machine Learning for Muscle Fatigue Estimation

by Yassine Bouteraa, Ismail Ben Abdallah and Khalil Boukthir

Bioengineering 2023, 10(2), 219; https://doi.org/10.3390/bioengineering10020219 - 6 Feb 2023

Cited by 9 | Viewed by 4881

Abstract

In this research, a new remote rehabilitation system was developed that integrates an IoT-based connected robot intended for wrist and forearm rehabilitation. In fact, the mathematical model of the wrist and forearm joints was developed and integrated into the main controller. The proposed new rehabilitation protocol consists of three main sessions: the first is dedicated to the extraction of the passive components of the dynamic model of wrist–forearm biomechanics while the active components are extracted in the second session. The third session consists of performing continuous exercises using the determined dynamic model of the forearm–wrist joints, taking into account the torque generated by muscle fatigue. The main objective of this protocol is to determine the state level of the affected wrist and above all to provide a dynamic model in which the torque generated by the robot and the torque supplied by the patient are combined, taking into account the constraints of fatigue. A Support Vector Machine (SVM) classifier is designed for the estimation of muscle fatigue based on the features extracted from the electromyography (EMG) signal acquired from the patient. The results show that the developed rehabilitation system allows a good progression of the joint’s range of motion as well as the resistive-active torques. Full article

(This article belongs to the Special Issue Data Processing and Machine Learning for Assistive and Rehabilitation Technologies)

► Show Figures

Figure 1

Other

Jump to: Research

18 pages, 1433 KiB

Open AccessSystematic Review

Artificial Intelligence Models for the Automation of Standard Diagnostics in Sleep Medicine—A Systematic Review

by Maha Alattar, Alok Govind and Shraddha Mainali

Bioengineering 2024, 11(3), 206; https://doi.org/10.3390/bioengineering11030206 - 22 Feb 2024

Cited by 2 | Viewed by 2252

Abstract

Sleep disorders, prevalent in the general population, present significant health challenges. The current diagnostic approach, based on a manual analysis of overnight polysomnograms (PSGs), is costly and time-consuming. Artificial intelligence has emerged as a promising tool in this context, offering a more accessible and personalized approach to diagnosis, particularly beneficial for under-served populations. This is a systematic review of AI-based models for sleep disorder diagnostics that were trained, validated, and tested on diverse clinical datasets. An extensive search of PubMed and IEEE databases yielded 2114 articles, but only 18 met our stringent selection criteria, underscoring the scarcity of thoroughly validated AI models in sleep medicine. The findings emphasize the necessity of a rigorous validation of AI models on multimodal clinical data, a step crucial for their integration into clinical practice. This would be in line with the American Academy of Sleep Medicine’s support of AI research. Full article

(This article belongs to the Special Issue Data Processing and Machine Learning for Assistive and Rehabilitation Technologies)

► Show Figures

Figure 1

22 pages, 1480 KiB

Open AccessSystematic Review

Neural Plasticity Changes Induced by Motor Robotic Rehabilitation in Stroke Patients: The Contribution of Functional Neuroimaging

by Lilla Bonanno, Antonio Cannuli, Loris Pignolo, Silvia Marino, Angelo Quartarone, Rocco Salvatore Calabrò and Antonio Cerasa

Bioengineering 2023, 10(8), 990; https://doi.org/10.3390/bioengineering10080990 - 21 Aug 2023

Cited by 7 | Viewed by 5812

Abstract

Robotic rehabilitation is one of the most advanced treatments helping people with stroke to faster recovery from motor deficits. The clinical impact of this type of treatment has been widely defined and established using clinical scales. The neurofunctional indicators of motor recovery following conventional rehabilitation treatments have already been identified by previous meta-analytic investigations. However, a clear definition of the neural correlates associated with robotic neurorehabilitation treatment has never been performed. This systematic review assesses the neurofunctional correlates (fMRI, fNIRS) of cutting-edge robotic therapies in enhancing motor recovery of stroke populations in accordance with PRISMA standards. A total of 7, of the initial yield of 150 articles, have been included in this review. Lessons from these studies suggest that neural plasticity within the ipsilateral primary motor cortex, the contralateral sensorimotor cortex, and the premotor cortices are more sensitive to compensation strategies reflecting upper and lower limbs’ motor recovery despite the high heterogeneity in robotic devices, clinical status, and neuroimaging procedures. Unfortunately, the paucity of RCT studies prevents us from understanding the neurobiological differences induced by robotic devices with respect to traditional rehabilitation approaches. Despite this technology dating to the early 1990s, there is a need to translate more functional neuroimaging markers in clinical settings since they provide a unique opportunity to examine, in-depth, the brain plasticity changes induced by robotic rehabilitation. Full article

(This article belongs to the Special Issue Data Processing and Machine Learning for Assistive and Rehabilitation Technologies)

► Show Figures

Journal Menu

Journal Browser

Data Processing and Machine Learning for Assistive and Rehabilitation Technologies

Share This Special Issue

Special Issue Editors

Special Issue Information

Keywords

Benefits of Publishing in a Special Issue

Published Papers (9 papers)

Research

Other

Further Information

Guidelines

MDPI Initiatives

Follow MDPI