Sparse learned kernels for interpretable and efficient medical time series processing
This addresses the need for rapid, reliable, and interpretable medical signal processing for clinical decision-making, particularly in consumer wearables, but is incremental as it builds on existing kernel and sparse network methods.
The authors tackled the problem of deep learning models being compute-intensive and lacking interpretability in medical time series processing by proposing Sparse Mixture of Learned Kernels (SMoLK), which matches the performance of much larger models while providing interpretability and efficiency.
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.