Sparse learned kernels for interpretable and efficient medical time series processing
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2023
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909336647761920 |
|---|---|
| author | Chen, Sully F. Guo, Zhicheng Ding, Cheng Hu, Xiao Rudin, Cynthia |
| author_facet | Chen, Sully F. Guo, Zhicheng Ding, Cheng Hu, Xiao Rudin, Cynthia |
| contents | Rapid, reliable, and accurate interpretation of medical time-series signals is crucial for high-stakes clinical decision-making. Deep learning methods offered unprecedented performance in medical signal processing but at a cost: they were compute-intensive and lacked interpretability. We propose Sparse Mixture of Learned Kernels (SMoLK), an interpretable architecture for medical time series processing. SMoLK learns a set of lightweight flexible kernels that form a single-layer sparse neural network, providing not only interpretability, but also efficiency, robustness, and generalization to unseen data distributions. We introduce a parameter reduction techniques to reduce the size of SMoLK's networks while maintaining performance. We test SMoLK on two important tasks common to many consumer wearables: photoplethysmography (PPG) artifact detection and atrial fibrillation detection from single-lead electrocardiograms (ECGs). We find that SMoLK matches the performance of models orders of magnitude larger. It is particularly suited for real-time applications using low-power devices, and its interpretability benefits high-stakes situations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2307_05385 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Sparse learned kernels for interpretable and efficient medical time series processing Chen, Sully F. Guo, Zhicheng Ding, Cheng Hu, Xiao Rudin, Cynthia Signal Processing Artificial Intelligence Machine Learning Rapid, reliable, and accurate interpretation of medical time-series signals is crucial for high-stakes clinical decision-making. Deep learning methods offered unprecedented performance in medical signal processing but at a cost: they were compute-intensive and lacked interpretability. We propose Sparse Mixture of Learned Kernels (SMoLK), an interpretable architecture for medical time series processing. SMoLK learns a set of lightweight flexible kernels that form a single-layer sparse neural network, providing not only interpretability, but also efficiency, robustness, and generalization to unseen data distributions. We introduce a parameter reduction techniques to reduce the size of SMoLK's networks while maintaining performance. We test SMoLK on two important tasks common to many consumer wearables: photoplethysmography (PPG) artifact detection and atrial fibrillation detection from single-lead electrocardiograms (ECGs). We find that SMoLK matches the performance of models orders of magnitude larger. It is particularly suited for real-time applications using low-power devices, and its interpretability benefits high-stakes situations. |
| title | Sparse learned kernels for interpretable and efficient medical time series processing |
| topic | Signal Processing Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2307.05385 |