dorsal/arxiv
View SchemaPareto-Optimal Model Selection for Low-Cost, Single-Lead EMG Control in Embedded Systems
| Authors | Carl Vincent Ladres Kho |
|---|---|
| Categories | |
| ArXiv ID | 2601.06516vv1 |
| URL | https://arxiv.org/abs/2601.06516 |
| License | http://creativecommons.org/licenses/by/4.0/ |
Abstract
Consumer-grade biosensors offer a cost-effective alternative to medical-grade electromyography (EMG) systems, reducing hardware costs from thousands of dollars to approximately $13. However, these low-cost sensors introduce significant signal instability and motion artifacts. Deploying machine learning models on resource-constrained edge devices like the ESP32 presents a challenge: balancing classification accuracy with strict latency (<100ms) and memory (<320KB) constraints. Using a single-subject dataset comprising 1,540 seconds of raw data (1.54M data points, segmented into ~1,300 one-second windows), I evaluate 18 model architectures, ranging from statistical heuristics to deep transfer learning (ResNet50) and custom hybrid networks (MaxCRNN). While my custom "MaxCRNN" (Inception + Bi-LSTM + Attention) achieved the highest safety (99% Precision) and robustness, I identify Random Forest (74% accuracy) as the Pareto-optimal solution for embedded control on legacy microcontrollers. I demonstrate that reliable, low-latency EMG control is feasible on commodity hardware, with Deep Learning offering a path to near-perfect reliability on modern Edge AI accelerators.
{
"annotation_id": "65513d1c-96ec-488c-9b04-80b1ded779a3",
"date_created": "2026-02-17T05:53:08.707000Z",
"date_modified": "2026-02-17T05:53:08.707000Z",
"file_hash": "7a26b37bc4c3b3031872332a2eabeb575ca13c938ee56fb4dbd677b4894426cd",
"private": false,
"record": {
"abstract": "Consumer-grade biosensors offer a cost-effective alternative to medical-grade electromyography (EMG) systems, reducing hardware costs from thousands of dollars to approximately $13. However, these low-cost sensors introduce significant signal instability and motion artifacts. Deploying machine learning models on resource-constrained edge devices like the ESP32 presents a challenge: balancing classification accuracy with strict latency (\u003c100ms) and memory (\u003c320KB) constraints. Using a single-subject dataset comprising 1,540 seconds of raw data (1.54M data points, segmented into ~1,300 one-second windows), I evaluate 18 model architectures, ranging from statistical heuristics to deep transfer learning (ResNet50) and custom hybrid networks (MaxCRNN). While my custom \"MaxCRNN\" (Inception + Bi-LSTM + Attention) achieved the highest safety (99% Precision) and robustness, I identify Random Forest (74% accuracy) as the Pareto-optimal solution for embedded control on legacy microcontrollers. I demonstrate that reliable, low-latency EMG control is feasible on commodity hardware, with Deep Learning offering a path to near-perfect reliability on modern Edge AI accelerators.",
"arxiv_id": "2601.06516",
"authors": [
"Carl Vincent Ladres Kho"
],
"categories": [
"cs.HC",
"cs.LG",
"eess.SP"
],
"license": "http://creativecommons.org/licenses/by/4.0/",
"title": "Pareto-Optimal Model Selection for Low-Cost, Single-Lead EMG Control in Embedded Systems",
"url": "https://arxiv.org/abs/2601.06516",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "339e0d16-64fa-4558-a0d6-93d4a8d29717",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}