Research map / Vision-language and multimodal / Human and embodied modeling
Clinical-BERT: Vision-Language Pre-training for Radiograph Diagnosis and Reports Generation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use Clinical-BERT rather than PPKED | Radiograph report generation · IU X-Ray | BLEU-1: 0.483 -> 0.495 | Helped |
| Use Clinical-BERT rather than ASGK | Radiograph report generation · COV-CTR | CIDEr: 0.684 -> 1.218 | Helped |
| Use Clinical-BERT rather than PPKED | Radiograph report generation · MIMIC-CXR | BLEU-1: 0.360 -> 0.383 | Helped |
| Use Clinical-BERT rather than DME | Radiograph diagnosis · NIH ChestX-ray14 | mean AUC: 0.8349 -> 0.8450 | Helped |
| Add Clinical Diagnosis pre-training (ablation) | Clinical-BERT report generation · IU X-Ray | BLEU-1: 0.465 -> 0.479 | Helped |
| Add Masked MeSH Modeling after Clinical Diagnosis (ablation) | Clinical-BERT report generation · IU X-Ray | BLEU-1: 0.479 -> 0.489 | Helped |
| Add Image-MeSH Matching after CD and MMM (ablation) | Clinical-BERT report generation · MIMIC-CXR | BLEU-1: 0.363 -> 0.372 | Helped |
| Add sparse attention to Image-MeSH Matching (ablation) | Clinical-BERT report generation · MIMIC-CXR | BLEU-1: 0.372 -> 0.383 | Helped |
| Add sparse attention to Image-MeSH Matching (ablation) | Clinical-BERT report generation · IU X-Ray | ROUGE-L: 0.371 -> 0.376 | Mixed |
| Increase sparse-attention K from 12 (ablation) | Clinical-BERT report generation · MIMIC-CXR | BLEU-1: 0.383 -> 0.378 | Hurt |
Closest papers by meaning
- Large-scale and Fine-grained Vision-language Pre-training for Enhanced CT Image Understanding (ICLR 2025 · similarity 0.87)
- RadZero: Similarity-Based Cross-Attention for Explainable Vision-Language Alignment in Chest X-ray with Zero-Shot Multi-Task Capability (NeurIPS 2025 · similarity 0.86)
- Unlocking the Power of Spatial and Temporal Information in Medical Multimodal Pre-training (ICML 2024 · similarity 0.86)
- HC-LLM: Historical-Constrained Large Language Models for Radiology Report Generation (AAAI 2025 · similarity 0.85)
- Bootstrapping Chest CT Image Understanding by Distilling Knowledge from X-ray Expert Models (CVPR 2024 · similarity 0.85)
- LLM-CXR: Instruction-Finetuned LLM for CXR Image Understanding and Generation (ICLR 2024 · similarity 0.85)
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.