Metraipolis › archive › dc_8bf850aaefee

Few-Shot Prototype Head Adaptation for On-Device ECG Personalization on PSoC~6 — arXiv:2610.06241

Source: arXiv
Original: https://arxiv.org/abs/2610.06241
License (as archived): CC0-1.0
Archived: 2026-10-06T05:27:32.539Z
SHA-256 of the archived text: 278764266d651cf38f08cdd0d0e35e3eaac48db350f7137d13d1cfda2a7d7057
Stable URL of this page: https://metraipolis.com/documento/dc_8bf850aaefee

How to cite

Few-Shot Prototype Head Adaptation for On-Device ECG Personalization on PSoC~6 — arXiv:2610.06241 — arXiv, https://arxiv.org/abs/2610.06241 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8bf850aaefee (Metraipolis archive dc_8bf850aaefee, sha256 278764266d651cf38f08cdd0d0e35e3eaac48db350f7137d13d1cfda2a7d7057)

BibTeX

@misc{dc8bf850aaefee,
  title = {Few-Shot Prototype Head Adaptation for On-Device ECG Personalization on PSoC~6 — arXiv:2610.06241},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.06241},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8bf850aaefee; sha256 278764266d651cf38f08cdd0d0e35e3eaac48db350f7137d13d1cfda2a7d7057},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.06241. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Wearable and bedside electrocardiogram (ECG) monitors must adapt to patient-specific morphology to maintain arrhythmia detection accuracy across users, yet personalization is typically performed offline and cannot account for individual physiology, electrode placement, or recording drift. On-device adaptation by backpropagation is expensive for microcontroller-class medical devices because it requires an optimizer state, repeated backward passes through convolutional layers, and labeled arrhythmic beats that may not be available at deployment time. This letter proposes prototype-only head adaptation as a compact personalization primitive for TinyML ECG systems. A one-dimensional convolutional neural network (1-D CNN; 1,314 parameters and 72.6k multiply-accumulate operations per beat) is trained offline on the MIT-BIH Arrhythmia Database under an inter-patient protocol, frozen as a feature extractor, and exported to a PSoC 6 microcontroller. Patient-specific adaptation then reduces to computing closed-form class means in a 32-dimensional embedding space, requiring no convolutional backward pass, no iterative optimization, and only one forward pass per support beat. Prototype adaptation improves inter-patient macro-F1 from 0.635/0.639/0.646 to 0.731/0.771/0.797 at 1/5/10-shot, outperforming linear stochastic-gradient-descent (SGD) head fine-tuning at every shot count for the target tiny backbone. On-device replay over 18 one-shot episodes on a PSoC 6 Cortex-M4F matches the host macro-F1 for the prototype head (0.798), with 11.39 ms per beat, 5.2 KB flash, and 22.2 KB SRAM. A restricted variant that updates only the normal-class prototype from passively buffered sinus beats yields a consistent +0.05 macro-F1 gain, reducing the annotation burden during initial

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.06241 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-06T05:24:06.135Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).