Special Issue "Machine Learning and Deep Learning for Biosignals Interpretation"
Deadline for manuscript submissions: 30 April 2021.
Interests: Semantic Web; semantic annotation; machine learning; artificial neural networks
Interests: EMG signal processing (filtering, feature extraction, pattern recognition, time–frequency analysis) and interpretation (physiology, clinics, sport); gait analysis; static and perturbed posturography; machine learning applications in motion analysis
Special Issues and Collections in MDPI journals
Interests: natural language processing; machine learning; text mining; web content extraction; social network analysis
The automatic or semi-automatic analysis and interpretation of biosignals is a prominent multidisciplinary research area encompassing biology, medicine, engineering, and computer science. The electrical, chemical, and mechanical activity occurring during a biological event often produces signals that can be measured and analyzed. Biosignals, therefore, could contain useful information to interpret the underlying physiological mechanisms of a specific biological event or system. The research field is gaining more and more momentum also thanks to advances in wearable sensors, which open new scenarios, e.g., for continuous monitoring of health-related parameters as well as to the advancements in deep learning, which has demonstrated its effectiveness in a variety of contexts, including health-related contexts.
While the traditional machine learning (ML) approaches to biosignal interpretation are based on hand-engineered features, in deep learning (DL)-based approaches, the signal’s features are automatically learned from “raw” data. However, ML and DL approaches are not opposed. On the one hand, biosignals are seldom used in their raw form but, rather, they usually require a preprocessing step, e.g., frequency filtering and signal smoothing, which can be considered as a sort of feature extraction. On the other hand, they both have strong and weak points, suggesting that hybrid approaches could do better in some tasks.
Machine learning approaches are generally considered better when little data is available for training and have demonstrated promising results in several applications. However, they usually lack generalization ability and might not be good in adapting to new situations and different conditions. Deep learning approaches have achieved better performances in a variety of applications, especially when huge amounts of data are available for learning. However, training a DL system can be very computationally expensive, making it difficult to integrate new data. Moreover, DL models are black boxes and usually do not provide easy ways to interpret the results, which can be of primary importance in a clinical context.
This Special Issue aims to collect relevant research advancements in biosignal processing and information/knowledge extraction from biosignals based on machine learning and deep learning, as well as to report the design and development of novel systems that use biosignals and ML/DL to address specific application scenarios, which include, but are not limited to the following:
- Analysis of human movements (e.g., gait event detection, gesture recognition, etc.)
- Emotion detection and classification
- Health status monitoring (e.g., sleep quality measurement, arrhythmia detection, etc.)
- Support for diagnosis and therapy management of pathologies
- Assistive devices
- Human–computer interaction
In the context of this Special Issue, relevant classes of biosignals include the following:
- Bioelectrical signals generated by nerves and muscle cells (e.g., EEG, EMG, ECG, EGG)
- Biomagnetic signals associated with specific physiological activity typically linked to an accompanying electric field from a specific tissue or organ (e.g., MEG, MNG, MCG)
- Biomechanical signals describing mechanical functions of biological systems including motion, displacement, tension, force, pressure and flow, production of measurable biological signals (e.g., kinematics, dynamics, blood pressure)
- Biochemical signals containing information about changes in concentration of chemical agents in the body
Prof. Dr. Christian Morbidoni
Prof. Dr. Francesco Di Nardo
Prof. Dr. Alessandro Cucchiarelli
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 papers will be 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 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 1500 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.
- human movement analysis (e.g., gait event detection, gesture recognition, etc.)
- emotion detection and classification
- health status monitoring (e.g., sleep quality measurement, arrhythmia detection, etc.)
- support for diagnosis and therapy management of pathologies
- assistive devices
- human–computer interaction