Profiling a Raspberry Pi-Based Motor Imagery Classification to Facilitate At-Home BCI for Children with Disabilities
Bibliographic record
Abstract
There has been incremental progress in moving BCI out of the laboratory environment and into the homes of those who would benefit most, especially children living with severe physical disabilities. Practical issues, such as available computational resources and long calibration times, have slowed down the adoption of such systems. To develop an efficient and scalable machine learning framework consistent with early approaches that facilitate at-home BCI use, this study provides valuable insights into measuring the behavioral characteristics of a Raspberry Pi 4 (RPi4) during the operation and execution of standard BCI processes, including the training and evaluation of classifier models. The results, which evaluated ten standard classifiers, including the Riemannian Geometry (RG) framework and more advanced deep learning approaches like Artificial Neural Network (ANN), were profiled on RPi4. These were compared to Desktop and MacBook computations for metrics such as training time, inference time, peak memory, and incremental memory usage, with computational bottlenecks identified. Our assessment revealed comparable performance metrics (84.3% of accuracy, recall, and f1_score, and 84.7% precision) for the neural network models despite the lower computational resources. Profiling results, including 1.74 sec training time, 0.405 sec inference time, 1154.9 MiB peak memory, and 405.2 MiB incremental memory usage, also demonstrated that the RPi4 is a potentially viable device for low-cost BCI systems. However, high-resource demanding classifiers such as ANN may need to be carefully considered in their implementation, which, in turn, will scale down the potential cost and complexity of adopting practical, impactful at-home BCI systems.
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 | 0.000 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
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".