Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | https://arxiv.org/abs/2503.05347 |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866913974878994432 |
|---|---|
| author | Zhang, Zhenxuan Lee, Kinhei Jing, Peiyuan Deng, Weihang Zhou, Huichi Jin, Zihao Huang, Jiahao Gao, Zhifan Marshall, Dominic C Fang, Yingying Yang, Guang |
| author_facet | Zhang, Zhenxuan Lee, Kinhei Jing, Peiyuan Deng, Weihang Zhou, Huichi Jin, Zihao Huang, Jiahao Gao, Zhifan Marshall, Dominic C Fang, Yingying Yang, Guang |
| contents | Automatic medical report generation has the potential to support clinical diagnosis, reduce the workload of radiologists, and demonstrate potential for enhancing diagnostic consistency. However, current evaluation metrics often fail to reflect the clinical reliability of generated reports. Early overlap-based methods focus on textual matches between predicted and ground-truth entities but miss fine-grained clinical details (e.g., anatomical location, severity). Some diagnostic metrics are limited by fixed vocabularies or templates, reducing their ability to capture diverse clinical expressions. LLM-based approaches further lack interpretable reasoning steps, making it hard to assess or trust their behavior in safety-critical settings. These limitations hinder the comprehensive assessment of the reliability of generated reports and pose risks in their selection for clinical use. Therefore, we propose a Granular Explainable Multi-Agent Score (GEMA-Score) in this paper, which conducts both objective quantification and subjective evaluation through a large language model-based multi-agent workflow. Our GEMA-Score parses structured reports and employs stable calculations through interactive exchanges of information among agents to assess disease diagnosis, location, severity, and uncertainty. Additionally, an LLM-based scoring agent evaluates completeness, readability, and clinical terminology while providing explanatory feedback. Extensive experiments validate that GEMA-Score achieves the highest correlation with human expert evaluations on a public dataset, demonstrating its effectiveness in clinical scoring (Kendall coefficient = $0.69$ for ReXVal dataset and Kendall coefficient = $0.45$ for RadEvalX dataset). The anonymous project demo is available at: https://github.com/Zhenxuan-Zhang/GEMA_score. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_05347 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | GEMA-Score: Granular Explainable Multi-Agent Scoring Framework for Radiology Report Evaluation Zhang, Zhenxuan Lee, Kinhei Jing, Peiyuan Deng, Weihang Zhou, Huichi Jin, Zihao Huang, Jiahao Gao, Zhifan Marshall, Dominic C Fang, Yingying Yang, Guang Computation and Language Multiagent Systems Automatic medical report generation has the potential to support clinical diagnosis, reduce the workload of radiologists, and demonstrate potential for enhancing diagnostic consistency. However, current evaluation metrics often fail to reflect the clinical reliability of generated reports. Early overlap-based methods focus on textual matches between predicted and ground-truth entities but miss fine-grained clinical details (e.g., anatomical location, severity). Some diagnostic metrics are limited by fixed vocabularies or templates, reducing their ability to capture diverse clinical expressions. LLM-based approaches further lack interpretable reasoning steps, making it hard to assess or trust their behavior in safety-critical settings. These limitations hinder the comprehensive assessment of the reliability of generated reports and pose risks in their selection for clinical use. Therefore, we propose a Granular Explainable Multi-Agent Score (GEMA-Score) in this paper, which conducts both objective quantification and subjective evaluation through a large language model-based multi-agent workflow. Our GEMA-Score parses structured reports and employs stable calculations through interactive exchanges of information among agents to assess disease diagnosis, location, severity, and uncertainty. Additionally, an LLM-based scoring agent evaluates completeness, readability, and clinical terminology while providing explanatory feedback. Extensive experiments validate that GEMA-Score achieves the highest correlation with human expert evaluations on a public dataset, demonstrating its effectiveness in clinical scoring (Kendall coefficient = $0.69$ for ReXVal dataset and Kendall coefficient = $0.45$ for RadEvalX dataset). The anonymous project demo is available at: https://github.com/Zhenxuan-Zhang/GEMA_score. |
| title | GEMA-Score: Granular Explainable Multi-Agent Scoring Framework for Radiology Report Evaluation |
| topic | Computation and Language Multiagent Systems |
| url | https://arxiv.org/abs/2503.05347 |