Human-Machine Interaction Using Discrete Myoelectric Control: Contrastive Learning Reduces False Activations During Activities of Daily Living
Bibliographic record
Abstract
Although myoelectric control has predominantly been used as a continuous input for prosthesis control, there are many applications in robotics, mixed reality, and wearable devices where discrete, event-driven inputs may be preferred. Furthermore, the traditional closed-set assumption in pros-thetics, where users are assumed to be constantly (and only) controlling the target device, may be undesirable for these emerging applications. To enable the real-world viability of such EMG-based inputs, myoelectric control research must move toward open-set systems that can reliably recognize and classify target gesture commands and discriminate them from out-of-set inputs. This work proposes and evaluates an end-to-end LSTM-based architecture that leverages contrastive learning to recognize a set of dynamic gestures while simultaneously rejecting activities of daily living. Compared to the current standard training approach (which generally uses the cross entropy loss function), the proposed contrastive approach significantly reduces the number of false positives (p<0.0005) during a set of activities of daily living (including walking, writing, typing, driving, and phone use) while maintaining high accuracy (95%) on the closed-set target gestures. These results highlight a promising path for developing discrete myoelectric control as an always-available human-machine interface.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".