DBGL: Decay-aware Bipartite Graph Learning for Irregular Medical Time Series Classification

arXiv cs.AI / 4/15/2026

💬 OpinionIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper proposes DBGL (Decay-aware Bipartite Graph Learning) to improve classification of irregular medical time series where heterogeneous sampling rates, asynchronous observations, and variable gaps introduce major modeling challenges.
It builds a patient-variable bipartite graph that captures irregular sampling without artificial time alignment and adaptively models variable relationships to better represent temporal sampling irregularity.
DBGL further introduces a node-specific temporal decay encoding mechanism that uses sampling intervals to model variable-specific decay rates, aiming to reflect true irregular temporal dynamics.
Experiments on four public datasets show DBGL achieves better performance than existing baseline methods, indicating improved representation learning for irregular clinical data.

Abstract

Irregular Medical Time Series play a critical role in the clinical domain to better understand the patient's condition. However, inherent irregularity arising from heterogeneous sampling rates, asynchronous observations, and variable gaps poses key challenges for reliable modeling. Existing methods often distort temporal sampling irregularity and missingness patterns while failing to capture variable decay irregularity, resulting in suboptimal representations. To address these limitations, we introduce DBGL, Decay-Aware Bipartite Graph Learning for Irregular Medical Time Series. DBGL first introduces a patient-variable bipartite graph that simultaneously captures irregular sampling patterns without artificial alignment and adaptively models variable relationships for temporal sampling irregularity modeling, enhancing representation learning. To model variable decay irregularity, DBGL designs a novel node-specific temporal decay encoding mechanism that captures each variable's decay rates based on sampling interval, yielding a more accurate and faithful representation of irregular temporal dynamics. We evaluate the performance of DBGL on four publicly available datasets, and the results show that DBGL outperforms all baselines.