RaTEScore: A Metric for Radiology Report Generation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917812972290048 |
|---|---|
| author | Zhao, Weike Wu, Chaoyi Zhang, Xiaoman Zhang, Ya Wang, Yanfeng Xie, Weidi |
| author_facet | Zhao, Weike Wu, Chaoyi Zhang, Xiaoman Zhang, Ya Wang, Yanfeng Xie, Weidi |
| contents | This paper introduces a novel, entity-aware metric, termed as Radiological Report (Text) Evaluation (RaTEScore), to assess the quality of medical reports generated by AI models. RaTEScore emphasizes crucial medical entities such as diagnostic outcomes and anatomical details, and is robust against complex medical synonyms and sensitive to negation expressions. Technically, we developed a comprehensive medical NER dataset, RaTE-NER, and trained an NER model specifically for this purpose. This model enables the decomposition of complex radiological reports into constituent medical entities. The metric itself is derived by comparing the similarity of entity embeddings, obtained from a language model, based on their types and relevance to clinical significance. Our evaluations demonstrate that RaTEScore aligns more closely with human preference than existing metrics, validated both on established public benchmarks and our newly proposed RaTE-Eval benchmark. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_16845 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | RaTEScore: A Metric for Radiology Report Generation Zhao, Weike Wu, Chaoyi Zhang, Xiaoman Zhang, Ya Wang, Yanfeng Xie, Weidi Computation and Language This paper introduces a novel, entity-aware metric, termed as Radiological Report (Text) Evaluation (RaTEScore), to assess the quality of medical reports generated by AI models. RaTEScore emphasizes crucial medical entities such as diagnostic outcomes and anatomical details, and is robust against complex medical synonyms and sensitive to negation expressions. Technically, we developed a comprehensive medical NER dataset, RaTE-NER, and trained an NER model specifically for this purpose. This model enables the decomposition of complex radiological reports into constituent medical entities. The metric itself is derived by comparing the similarity of entity embeddings, obtained from a language model, based on their types and relevance to clinical significance. Our evaluations demonstrate that RaTEScore aligns more closely with human preference than existing metrics, validated both on established public benchmarks and our newly proposed RaTE-Eval benchmark. |
| title | RaTEScore: A Metric for Radiology Report Generation |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2406.16845 |