Research map / Label-efficient and robust learning / Recognition and long-tail
Towards Semi-Structured Automatic ICD Coding via Tree-based Contrastive Learning
Chang Lu, Chandan K. Reddy, Ping Wang, Yue Ning
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 contrastive pre-training and masked section training (CM) | MSMN · MIMIC-50 | Macro F1: 68.1 -> 69.1 | Improved |
| add CM | MultiResCNN · MIMIC-50 | Macro F1: 60.8 -> 62.2 | Improved |
| add CM | PLM-ICD · MIMIC-50 | Macro F1: 64.5 -> 65.2 | Improved |
| add CM | MultiResCNN · MIMIC-rare-50 | Macro F1: 11.2 -> 22.8 | Improved |
| add CM | Hierarchical · MIMIC-rare-50 | Macro F1: 23.1 -> 32.0 | Improved |
| add CM | MSMN · MIMIC-rare-50 | Macro F1: 23.7 -> 31.2 | Improved |
| add CM | MSMN · MIMIC-full | Macro F1: 10.3 -> 11.4 | Improved |
| add CM | PLM-ICD · MIMIC-full | Macro F1: 10.4 -> 10.6 | No clear effect |
| add CM | PLM-ICD · MIMIC-full | Micro F1: 59.8 -> 60.0 | No clear effect |
| use only contrastive pre-training with Jaccard similarity instead of tree edit distance (ablation) | JointLAAT, EffectiveCAN, MSMN · MIMIC-50 | performance | Hurt |
| use tree edit distance instead of Jaccard similarity for contrastive pre-training (ablation) | JointLAAT, EffectiveCAN, MSMN · MIMIC-50 | performance | Improved |
| add only masked section training versus only contrastive pre-training (ablation) | backbone ICD models · MIMIC-50 | performance | Mixed |
Closest papers by meaning
- Multi-Label Few-Shot ICD Coding as Autoregressive Generation with Prompt (AAAI 2023 · similarity 0.89)
- Bringing CLIP to the Clinic: Dynamic Soft Labels and Negation-Aware Learning for Medical Analysis (CVPR 2025 · similarity 0.83)
- Memorize and Rank: Elevating Large Language Models for Clinical Diagnosis Prediction (AAAI 2025 · similarity 0.81)
- MLIP: Enhancing Medical Visual Representation with Divergence Encoder and Knowledge-guided Contrastive Learning (CVPR 2024 · similarity 0.80)
- CUPCase: Clinically Uncommon Patient Cases and Diagnoses Dataset (AAAI 2025 · similarity 0.80)
- Multimodal Medical Code Tokenizer (ICML 2025 · 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.
