Machine-learning myoelectric prosthesis control - towards the advancement of assessing functional use and control skill

Aug 3, 2026

Upload by Himbo

The thesis advances methods for evaluating functional use and control skill in myoelectric hand prostheses, focusing on why standardized clinical tests fail to mirror daily user challenges. It argues that users select the path of least resistance and that embodiment includes intuitive assessment of control and mechanical failure risks. The work examines how intuitive control depends on stable input-output mappings, how EMG feature space metrics predict offline—not real-time—performance, and how increased technological complexity demands more complex testing paradigms.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages2

Reading Time5 min

AccessFree to download

Propositions

Daily-life control choices and embodiment

Limits of clinical testing and mapping stability

EMG metrics and real-time performance

Sensory challenges and complexity of evaluation

University of Groningen Machine-learning myoelectric prosthesis control: towards the advancement of assessing functional use and control skill Franzke, Andreas DOI: 10.33612/diss.663805052 IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below. Document Version Publisher's PDF, also known as Version of record Publication date: 2023 Link to publication in University of Groningen/UMCG research database Citation for published version (APA): Franzke, A. (2023) . Machine-learning myoelectric prosthesis control: towards the advancement of assessing functional use and control skill. [Thesis fully internal (DIV), University of Groningen] . University of Groningen. [https://doi.org/10.33612/diss.663805052](https://doi.org/10.33612/diss.663805052) Copyright Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons) . The publication may also be distributed here under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license. More information can be found on the University of Groningen website: [https://www.rug.nl/library/open-access/self-archiving-pure/taverne](https://www.rug.nl/library/open-access/self-archiving-pure/taverne)amendment. Take-down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. Downloaded from the University of Groningen/U MCG research database (Pure): [http://www.rug. nl/research/portal. For technical reasons the](http://www.rug. nl/research/portal. For technical reasons the)[ ](http://www.rug. nl/research/portal. For technical reasons the)[number of authors shown on this cover page is limited to 10 maximum.](number of authors shown on this cover page is limited to 10 maximum.) Download date: 02-08-2026 Propositions 1. To accomplish (motor) tasks in daily life, prosthesis users will choose the path of least resistance – irrespective of whether that actively involves the prosthesis or not (this thesis) . 2. The “embodiment” of a prosthetic device also involves a deep and intuitive perception of the user about the chance of control failure and mechanical failure of the device (this thesis) . 3. Multi-function prosthetic hands are mostly used for stabilization and support during bimanual tasks (this thesis) . 4. The majority of the standardized clinical tests for hand prostheses avoid situations which challenge the (myoelectric) prosthetic systems whereas prosthesis users encounter such situations every day (this thesis) . 5. An increase in the (technological) complexity of hand prostheses might require an increase in the complexity of the test for such devices as well (this thesis) . 6. Intuitive control doesn’t mean control without practice, errors and learning – intuitive control means that there is a clear and stable mapping between user input and motor output, where every error provides the user with useful information about that mapping (this thesis) . 7. EMG feature space metrics such as the separability index have proven to predict the (offline) performance of machine-learning algorithms. However, so far, they poorly predict the real-time control performance of a human (this thesis) . 8. The real challenge in achieving dexterous motor control with a hand prosthesis isnot in the translation of muscle signals to prosthetic motion, but in the lack of sensory structures in individuals with a missing hand (Inspired by Gesslbauer et al., 2017) . 9. Healthy biological systems are not robust to errors, but they are “anti-fragile”, meaning they thrive under (a certain degree of ) variation and errors. (Inspired by Taleb,

FAQs

Why do standardized clinical tests not reflect real-world myoelectric prosthesis use?

Most standardized hand prosthesis tests avoid situations that challenge myoelectric systems, even though prosthesis users face such situations every day.

What does “intuitive control” mean in the thesis?

Intuitive control does not eliminate practice, errors, and learning; it requires a clear and stable mapping between user input and motor output, where errors provide useful information about that mapping.

Why do EMG feature space metrics work well offline but not in real time?

Metrics such as separability index can predict offline performance of machine-learning algorithms, but they poorly predict real-time control performance of a human.

Aug 3, 2026

Upload by Himbo