RAxSS: Retrieval-Augmented Sparse Sampling for Explainable Variable-Length Medical Time Series Classification
This work addresses the problem of reliable and explainable classification of variable-length medical time series for clinical practitioners, representing an incremental improvement over existing sparse sampling and retrieval-augmented approaches.
The researchers tackled the challenge of classifying variable-length medical time series by developing a retrieval-augmented sparse sampling method that improves explainability and robustness to noise. Their approach achieved competitive iEEG classification performance across recordings from four medical centers, providing practitioners with greater transparency through an explicit evidence trail.
Medical time series analysis is challenging due to data sparsity, noise, and highly variable recording lengths. Prior work has shown that stochastic sparse sampling effectively handles variable-length signals, while retrieval-augmented approaches improve explainability and robustness to noise and weak temporal correlations. In this study, we generalize the stochastic sparse sampling framework for retrieval-informed classification. Specifically, we weight window predictions by within-channel similarity and aggregate them in probability space, yielding convex series-level scores and an explicit evidence trail for explainability. Our method achieves competitive iEEG classification performance and provides practitioners with greater transparency and explainability. We evaluate our method in iEEG recordings collected in four medical centers, demonstrating its potential for reliable and explainable clinical variable-length time series classification.