Learning ECG Signal Features Without Backpropagation Using Linear Laws
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2023
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911422793908224 |
|---|---|
| author | Pósfay, Péter Kurbucz, Marcell T. Kovács, Péter Jakovác, Antal |
| author_facet | Pósfay, Péter Kurbucz, Marcell T. Kovács, Péter Jakovác, Antal |
| contents | This paper introduces LLT-ECG, a novel method for electrocardiogram (ECG) signal classification that leverages concepts from theoretical physics to automatically generate features from time series data. Unlike traditional deep learning approaches, LLT-ECG operates in a forward manner, eliminating the need for backpropagation and hyperparameter tuning. By identifying linear laws that capture shared patterns within specific classes, the proposed method constructs a compact and verifiable representation, enhancing the effectiveness of downstream classifiers. We demonstrate LLT-ECG's state-of-the-art performance on real-world ECG datasets from PhysioNet, underscoring its potential for medical applications where speed and verifiability are crucial. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2307_01930 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Learning ECG Signal Features Without Backpropagation Using Linear Laws Pósfay, Péter Kurbucz, Marcell T. Kovács, Péter Jakovác, Antal Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition Applications 62H30, 68T10, 62M10, 92C50 J.3; I.5; I.2.0; G.3 This paper introduces LLT-ECG, a novel method for electrocardiogram (ECG) signal classification that leverages concepts from theoretical physics to automatically generate features from time series data. Unlike traditional deep learning approaches, LLT-ECG operates in a forward manner, eliminating the need for backpropagation and hyperparameter tuning. By identifying linear laws that capture shared patterns within specific classes, the proposed method constructs a compact and verifiable representation, enhancing the effectiveness of downstream classifiers. We demonstrate LLT-ECG's state-of-the-art performance on real-world ECG datasets from PhysioNet, underscoring its potential for medical applications where speed and verifiability are crucial. |
| title | Learning ECG Signal Features Without Backpropagation Using Linear Laws |
| topic | Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition Applications 62H30, 68T10, 62M10, 92C50 J.3; I.5; I.2.0; G.3 |
| url | https://arxiv.org/abs/2307.01930 |