A Tiny Transformer for Low-Power Arrhythmia Classification on Microcontrollers
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917699942088704 |
|---|---|
| author | Busia, Paola Scrugli, Matteo Antonio Jung, Victor Jean-Baptiste Benini, Luca Meloni, Paolo |
| author_facet | Busia, Paola Scrugli, Matteo Antonio Jung, Victor Jean-Baptiste Benini, Luca Meloni, Paolo |
| contents | Wearable systems for the continuous and real-time monitoring of cardiovascular diseases are becoming widespread and valuable assets in diagnosis and therapy. A promising approach for real-time analysis of the electrocardiographic (ECG) signal and the detection of heart conditions, such as arrhythmia, is represented by the transformer machine learning model. Transformers are powerful models for the classification of time series, although efficient implementation in the wearable domain raises significant design challenges, to combine adequate accuracy and a suitable complexity. In this work, we present a tiny transformer model for the analysis of the ECG signal, requiring only 6k parameters and reaching 98.97% accuracy in the recognition of the 5 most common arrhythmia classes from the MIT-BIH Arrhythmia database, assessed considering 8-bit integer inference as required for efficient execution on low-power microcontroller-based devices. We explored an augmentation-based training approach for improving the robustness against electrode motion artifacts noise, resulting in a worst-case post-deployment performance assessment of 98.36% accuracy. Suitability for wearable monitoring solutions is finally demonstrated through efficient deployment on the parallel ultra-low-power GAP9 processor, where inference execution requires 4.28ms and 0.09mJ. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_10748 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | A Tiny Transformer for Low-Power Arrhythmia Classification on Microcontrollers Busia, Paola Scrugli, Matteo Antonio Jung, Victor Jean-Baptiste Benini, Luca Meloni, Paolo Signal Processing Human-Computer Interaction Machine Learning Wearable systems for the continuous and real-time monitoring of cardiovascular diseases are becoming widespread and valuable assets in diagnosis and therapy. A promising approach for real-time analysis of the electrocardiographic (ECG) signal and the detection of heart conditions, such as arrhythmia, is represented by the transformer machine learning model. Transformers are powerful models for the classification of time series, although efficient implementation in the wearable domain raises significant design challenges, to combine adequate accuracy and a suitable complexity. In this work, we present a tiny transformer model for the analysis of the ECG signal, requiring only 6k parameters and reaching 98.97% accuracy in the recognition of the 5 most common arrhythmia classes from the MIT-BIH Arrhythmia database, assessed considering 8-bit integer inference as required for efficient execution on low-power microcontroller-based devices. We explored an augmentation-based training approach for improving the robustness against electrode motion artifacts noise, resulting in a worst-case post-deployment performance assessment of 98.36% accuracy. Suitability for wearable monitoring solutions is finally demonstrated through efficient deployment on the parallel ultra-low-power GAP9 processor, where inference execution requires 4.28ms and 0.09mJ. |
| title | A Tiny Transformer for Low-Power Arrhythmia Classification on Microcontrollers |
| topic | Signal Processing Human-Computer Interaction Machine Learning |
| url | https://arxiv.org/abs/2402.10748 |