Research map / Probabilistic and scientific modeling / Causal discovery
Building Causal Graphs from Medical Literature and Electronic Medical Records
Galia Nordon, Gideon Koren, Varda Shalev, Benny Kimelfeld, Uri Shalit, Kira Radinsky
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 |
|---|---|---|---|
| restrict nodes via embedding proximity instead of full SemCause graph | causal graph · Celiac | PPV: 0% -> 32% | Improved |
| restrict nodes via embedding proximity instead of full SemCause graph | causal graph · Atopic Dermatitis | PPV: 0% -> 24% | Improved |
| add EMR chi-square filtering to node2vec + co-occurrence graph (ablation) | causal graph · Celiac | PPV: 32% -> 50% | Improved |
| add EMR filtering to co-occurrence graph (ablation) | causal graph · Celiac | PPV: 24% -> 35% | Improved |
| add EMR filtering to word2vec graph (ablation) | causal graph · Celiac | PPV: 23% -> 27% | Improved |
| add EMR filtering to co-occurrence graph (ablation) | causal graph · Atopic Dermatitis | PPV: 7% -> 0% | Hurt |
| add EMR filtering to node2vec + co-occurrence graph (ablation) | causal graph · Atopic Dermatitis | PPV: 8% -> 0% | Hurt |
| add EMR filtering to node2vec + word2vec graph (ablation) | causal graph · Atopic Dermatitis | PPV: 24% -> 33% | Improved |
| add EMR filtering to node2vec combined graph (ablation) | causal graph · Atopic Dermatitis | PPV: 29% -> 25% | Hurt |
| add EMR filtering across node-selection methods (ablation) | causal graph · Celiac and Atopic Dermatitis | PPV | Mixed |
| use node2vec-based node selection instead of word2vec/co-occurrence | causal graph · Celiac | PPV: 24% -> 32% | Improved |
| stricter text edge selection (C2-C5 vs C1) (ablation) | causal graph · Celiac, word2vec nodes | PPV: 0.75 -> 0.05 | Hurt |
Closest papers by meaning
- End-To-End Causal Effect Estimation from Unstructured Natural Language Data (NeurIPS 2024 · similarity 0.85)
- DoAtlas-1: A Causal Compilation Paradigm for Clinical AI (NeurIPS 2026 · similarity 0.85)
- Identifying Macro Conditional Independencies and Macro Total Effects in Summary Causal Graphs with Latent Confounding (AAAI 2025 · similarity 0.85)
- Graph Distance Based on Cause-Effect Estimands with Latents (NeurIPS 2026 · similarity 0.84)
- Knowledge-Empowered Dynamic Graph Network for Irregularly Sampled Medical Time Series (NeurIPS 2024 · similarity 0.84)
- Local Identifying Causal Relations in the Presence of Latent Variables (ICML 2025 · similarity 0.84)
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.
