Machine Learning and Neuromorphic Computing to Improve Design, Usability, and Control of Smart Limb Prostheses

Dear Colleagues,

Recent advancements in mechatronics are enabling the development of highly advanced myoelectric prostheses. Mechanical design makes them capable of performing most of the tasks a human limb can do. Nevertheless, although these prostheses are more effective than traditional solutions such as aesthetic and body-powered prostheses, they are still rarely employed. 

The main reason of disaffection lies in the gap between the behavior of the device and the users’ commands, which generates frustration on the users.  Improving usability requires both working on better mechanical design and controllers and improving the user interfaces of the prosthetic devices. Other reasons are the long time required for training, the heavy weight of motorized prostheses, the need for frequent calibration, and the high costs, often not affordable for potential users.

While several methods have been developed in laboratory settings for controlling prostheses, those that are commercially available are much less advanced. Often they are difficult to control, and the effort spent is not compensated by the functionalities the prosthesis is able to perform.

To improve user acceptance, controllers able to interpret biological signals and to adapt to user needs are often based on the pattern recognition methods of the sEMG signals. Two recent techniques aimed at efficiently using the available biological data are seldom applied to prostheses: Machine Learning  (ML) and Neuromorphic Computing (NC).

Data-driven methods, as the ones of ML and Deep Learning, are powerful in the classification and prediction of biological signals. They offer extended functionalities in feature extraction and movement recognition. Moreover, they can work on various kinds of data, time series as well as numeric or symbolic data, making it feasible to move from EMG signals to many more biological signals.

Novel algorithms and hardware, such as the ones of NC, can accompany the ML methods and improve their computation performance. While NC hardware imitates the computing units and methods of the brain, NC develops in a broad way novel techniques for processing data inspired from biological processes.

This Special Issue intends to cover how a new design of the control system can address the above-mentioned challenges, providing smart and affordable prostheses .

In this Special Issue, original research articles and reviews are welcome.

Research areas may include (but are not limited to) the following five main areas:

1. Acquisition, interpretation, and classification of biological signals (EMG, visual data, etc.);
2. Sensory stimulation and feedback to the user;
3. Controllers and non-invasive user interfaces of prostheses with many degrees of freedom;
4. Neuroprostheses and neuromorphic sensors;
5. Mechanical and sensory–actuation design of prostheses,  covering innovative approaches in lightweight structures, efficient actuators, ergonomic architectures, and integrated sensory systems that complement advanced control algorithms. 

We look forward to receiving your contributions.

Prof. Giuseppina Gini

Prof. Michele Folgheraiter

Guest Editors

Prof. Dr. Giuseppina Gini
Prof. Dr. Michele Folgheraiter
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. Prosthesis 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 1600 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.

• machine learning
• neuromorphic computing
• smart limb prostheses
• advanced myoelectric prostheses
• sEMG signals
• mechanical and sensory–actuation design of prostheses