Chain-of-Influence: Tracing Interdependencies Across Time and Features in Clinical Predictive Modelings
This addresses the challenge of interpretability in clinical predictive modeling for healthcare professionals, offering transparency into temporal and cross-feature dependencies, though it is incremental as it builds on attention mechanisms.
The paper tackled the problem of capturing latent, time-varying dependencies in clinical time-series data by proposing Chain-of-Influence (CoI), an interpretable deep learning framework that constructs explicit graphs of feature interactions, and it significantly outperformed existing methods in predictive accuracy on mortality and disease progression tasks using datasets like MIMIC-IV.
Modeling clinical time-series data is hampered by the challenge of capturing latent, time-varying dependencies among features. State-of-the-art approaches often rely on black-box mechanisms or simple aggregation, failing to explicitly model how the influence of one clinical variable propagates through others over time. We propose $\textbf{Chain-of-Influence (CoI)}$, an interpretable deep learning framework that constructs an explicit, time-unfolded graph of feature interactions. CoI leverages a multi-level attention architecture: first, a temporal attention layer identifies critical time points in a patient's record; second, a cross-feature attention layer models the directed influence from features at these time points to subsequent features. This design enables the tracing of influence pathways, providing a granular audit trail that shows how any feature at any time contributes to the final prediction, both directly and through its influence on other variables. We evaluate CoI on mortality and disease progression tasks using the MIMIC-IV dataset and a private chronic kidney disease cohort. Our framework significantly outperforms existing methods in predictive accuracy. More importantly, through case studies, we show that CoI can uncover clinically meaningful, patient-specific patterns of disease progression that are opaque to other models, offering unprecedented transparency into the temporal and cross-feature dependencies that inform clinical decision-making.