{"slug": "ct-heg-a-bidirectional-timestamp-attributed-event-graph-for-icu-in-hospital-an", "title": "CT-HEG: A Bidirectional, Timestamp-Attributed Event Graph for ICU In-Hospital Mortality Prediction - An Architectural Ablation Study", "summary": "Researchers introduced CT-HEG, a bidirectional, timestamp-attributed event graph for ICU in-hospital mortality prediction, and their CHIRP-Net model achieved a 5-seed mean AUROC of 0.8449±0.0071 on MIMIC-IV v3.1 (31,142 ICU stays, 13.4% mortality). The study found that removing reverse edges cut AUROC by 0.1968±0.0073, and collapsing heterogeneous edge types into one relation outperformed the full model on all seeds. The authors caution that external validation, temporal evaluation, and fairness audit are needed before claiming robustness.", "body_md": "arXiv:2608.02663v1 Announce Type: new\nAbstract: Accurate ICU mortality prediction requires modeling irregular clinical observations across heterogeneous entity types. Existing sequence models handle irregular sampling but ignore typed relational structure; existing graph models assume fixed-interval inputs. We introduce the Continuous-Time Heterogeneous EHR Graph (CT-HEG) schema and evaluate which architectural choices drive predictive performance. CT-HEG encodes each ICU stay as a typed, timestamped graph with three node types (visit, vital, lab_event) and 2D edge attributes (t_hours/48, value_norm) encoding timing and value without imputation. We instantiate CT-HEG as CHIRP-Net, a four-layer heterogeneous GATv2Conv network, evaluated on MIMIC-IV v3.1 (31,142 ICU stays, LOS>=48h, 13.4% mortality) with five seeds and bootstrapped confidence intervals, against logistic regression, mTAND, a Transformer, and GRU-D, plus an ablation study. CHIRP-Net achieved 5-seed mean AUROC 0.8449+/-0.0071 (AUPRC 0.4958+/-0.0209); the ensemble achieved AUROC 0.8618 (95% CI: 0.8485-0.8745). Removing reverse edges disconnected observation nodes from the visit readout, cutting AUROC by 0.1968+/-0.0073. Time-attentive edge features contributed 0.0247+/-0.0093 AUROC. Collapsing heterogeneous edge types into one relation (7x fewer parameters) outperformed the full model on all seeds. Post-calibration ECE was 0.0307. Temporal and demographic subgroup analyses were explored but not reported here, pending follow-up work. Bidirectional connectivity was necessary for the model to use its inputs at all, and CT-HEG was reasonably well calibrated after validation-fitted temperature scaling. These results support CT-HEG for irregular EHR data, while external validation, a pre-specified temporal evaluation, and a demographic fairness audit remain necessary before any claim of robustness. Code: https://github.com/nasiruddinstudents-ctrl/chirp-net-mimic-iv.", "url": "https://wpnews.pro/news/ct-heg-a-bidirectional-timestamp-attributed-event-graph-for-icu-in-hospital-an", "canonical_source": "https://arxiv.org/abs/2608.02663", "published_at": "2026-08-05 04:00:00+00:00", "updated_at": "2026-08-05 04:01:51.355356+00:00", "lang": "en", "topics": ["machine-learning", "artificial-intelligence"], "entities": ["CT-HEG", "CHIRP-Net", "MIMIC-IV v3.1", "GATv2Conv", "mTAND", "GRU-D"], "alternates": {"html": "https://wpnews.pro/news/ct-heg-a-bidirectional-timestamp-attributed-event-graph-for-icu-in-hospital-an", "markdown": "https://wpnews.pro/news/ct-heg-a-bidirectional-timestamp-attributed-event-graph-for-icu-in-hospital-an.md", "text": "https://wpnews.pro/news/ct-heg-a-bidirectional-timestamp-attributed-event-graph-for-icu-in-hospital-an.txt", "jsonld": "https://wpnews.pro/news/ct-heg-a-bidirectional-timestamp-attributed-event-graph-for-icu-in-hospital-an.jsonld"}}