Back to feed
arXiv cs.LG
arXiv cs.LG
8/5/2026
CT-HEG: A Bidirectional, Timestamp-Attributed Event Graph for ICU In-Hospital Mortality Prediction - An Architectural Ablation Study

CT-HEG: A Bidirectional, Timestamp-Attributed Event Graph for ICU In-Hospital Mortality Prediction - An Architectural Ablation Study

Short summary

CT-HEG introduces a continuous-time heterogeneous graph schema for ICU mortality prediction, encoding each ICU stay as a typed, timestamped graph with visit, vital, and lab_event nodes. The CHIRP-Net model (4-layer GATv2Conv) achieves AUROC 0.8449 on MIMIC-IV, with bidirectional connectivity and time-attentive edge features being critical architectural choices. Ablation reveals that collapsing heterogeneous edge types into one relation with 7x fewer parameters actually outperforms the full model, and external validation plus fairness audits remain necessary before clinical deployment.

  • •CT-HEG encodes ICU stays as typed timestamped graphs with 2D edge attributes for timing and value
  • •CHIRP-Net achieves AUROC 0.8449 on MIMIC-IV (31K ICU stays), ensemble reaches 0.8618
  • •Bidirectional connectivity is essential; collapsing edge types with 7x fewer params outperforms full model

Generated with AI, which can make mistakes.

#research-breakthrough#open-source
Read full article at arXiv cs.LG

Is this a good recommendation for you?

Comments

Failed to load comments. Please try again.

Explore more

LLM4EHR: Aligning Clinical Time Series with Medical Event Sequences via Large Language Models
arXiv cs.LG
Large Language Models as Unified Multimodal Learners for Clinical Prediction
arXiv cs.CL
Domain Knowledge Based Temporal-Spatial Graph Convolution Network for ECG Recognition
arXiv cs.LG
SHIFT: Survival Prediction from Incomplete and Heterogeneous Genomic Data
arXiv cs.LG