ReTAMamba: Reliability-Aware Temporal Aggregation with Mamba for Irregular Clinical Time Series Prediction

TL;DR

ReTAMamba is a novel method that models irregular clinical time series by estimating observation reliability and integrating multi-resolution temporal information, leading to improved prediction accuracy.

Contribution

It introduces a reliability-aware temporal aggregation framework that reconstructs time series as token sequences and effectively captures temporal dynamics in irregular data.

Findings

ReTAMamba outperforms strong baselines on MIMIC-IV, eICU, and PhysioNet 2012 datasets.

It achieves average relative AUPRC gains of 7.51%, 7.80%, and 10.15%.

Learned decay rates are larger for dynamic signals, indicating effective modeling of information freshness.

Abstract

Clinical time-series data are difficult to model with methods designed for regular sequences because they exhibit irregular sampling, frequent missing values, and heterogeneous observation patterns across variables. Existing approaches commonly use observation masks and time-gap information, but they do not continuously capture the decaying reliability of past observations or consistently organize multi-resolution information within a coherent temporal context during aggregation. To address these limitations, we propose Reliability-aware Temporal Aggregation with Mamba (ReTAMamba), which reconstructs clinical time series as time-variable token sequences, estimates observation reliability from missingness and elapsed time, and augments interval summaries with statistical descriptors. Chronological Weaving is used to integrate short- and long-term temporal information within a coherent temporal context, and a budgeted token router is applied to constrain sequence length while preserving informative summaries. Experiments on MIMIC-IV, eICU, and PhysioNet 2012 show that ReTAMamba consistently improves AUPRC over strong baselines, with average relative gains of 7.51%, 7.80%, and 10.15%, respectively. Cohort-level and patient-level analyses on eICU further showed that the learned mean decay for more dynamic signals, such as heart rate and blood pressure, was 24.3% larger than that for relatively static signals, such as laboratory test variables. These findings suggest that effective prediction in irregular clinical time series requires modeling not only what was measured, but also when and how it was observed, including information freshness and observation timeliness.