Research map / Probabilistic and scientific modeling / Causal discovery
Debiased, Longitudinal and Coordinated Drug Recommendation through Multi-Visit Clinic Records
Hongda Sun, Shufang Xie, Shuqi Li, Yuhan Chen, Ji-Rong Wen, Rui Yan
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add symptom information (ablation) | DrugRec (w/o sypt.) · MIMIC-III | Jaccard: 0.5092 -> 0.5142 | Improved |
| add causal view of single-visit effect (ablation) | DrugRec (w/o Late) · MIMIC-III | Jaccard: 0.5142 -> 0.5167 | Improved |
| add multi-visit modeling (k=2) (ablation) | DrugRec (single) · MIMIC-III | Jaccard: 0.5167 -> 0.5220 | Improved |
| add multi-visit modeling (k=2) (ablation) | DrugRec (single) · MIMIC-III | PRAUC: 0.7667 -> 0.7720 | Improved |
| remove 2-SAT DDI control (ablation) | DrugRec · MIMIC-III | DDI Rate: 0.0597 -> 0.0774 | Hurt |
| remove 2-SAT DDI control (ablation) | DrugRec · MIMIC-III | Avg. # of Drugs: 22.0006 -> 23.9584 | Hurt |
| remove 2-SAT DDI control (ablation) | DrugRec · MIMIC-III | Jaccard: 0.5220 -> 0.5240 | No clear effect |
| remove symptom information (ablation) | DrugRec · MIMIC-III | F1: 0.6771 -> 0.6662 | Hurt |
| DrugRec-k (past visits) vs DrugRec-a (compressed all visits) | DrugRec · MIMIC-III | Jaccard: 0.5196 -> 0.5220 | Improved |
| DrugRec-k vs DrugRec-a | DrugRec · MIMIC-IV | Jaccard: 0.4162 -> 0.4194 | Improved |
| k from 0 to 1 past visits (ablation) | DrugRec-k · MIMIC-III | Jaccard: 0.5167 -> 0.5203 | Improved |
| k from 1 to 2 past visits (ablation) | DrugRec-k · MIMIC-III | Jaccard: 0.5203 -> 0.5220 | Improved |
Closest papers by meaning
- HI-DR: Exploiting Health Status-Aware Attention and an EHR Graph+ for Effective Medication Recommendation (AAAI 2025 · similarity 0.85)
- Fine-grained List-wise Alignment for Generative Medication Recommendation (NeurIPS 2025 · similarity 0.84)
- VITA: ‘Carefully Chosen and Weighted Less’ Is Better in Medication Recommendation (AAAI 2024 · similarity 0.82)
- PharmacoMatch: Efficient 3D Pharmacophore Screening via Neural Subgraph Matching (ICLR 2025 · similarity 0.81)
- SAFER: A Calibrated Risk-Aware Multimodal Recommendation Model for Dynamic Treatment Regimes (ICML 2025 · similarity 0.81)
- Risk Horizons: Structured Hypothesis Spaces for Longitudinal Clinical Prediction (NeurIPS 2026 · similarity 0.80)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
