Comparative Analysis of Abstractive Summarization Models for Clinical Radiology Reports
This work addresses the need for automated summarization to assist medical professionals in healthcare by providing a comparative analysis of existing models, but it is incremental as it applies standard methods to a specific domain without introducing new techniques.
This study compared several abstractive summarization models, including T5-base, BART-base, PEGASUS-x-base, ChatGPT-4, LLaMA-3-8B, and a custom Pointer Generator Network, to generate concise impressions from detailed radiology reports using the MIMIC-CXR dataset, evaluating them with metrics like ROUGE and BERTScore to identify strengths and limitations.
The findings section of a radiology report is often detailed and lengthy, whereas the impression section is comparatively more compact and captures key diagnostic conclusions. This research explores the use of advanced abstractive summarization models to generate the concise impression from the findings section of a radiology report. We have used the publicly available MIMIC-CXR dataset. A comparative analysis is conducted on leading pre-trained and open-source large language models, including T5-base, BART-base, PEGASUS-x-base, ChatGPT-4, LLaMA-3-8B, and a custom Pointer Generator Network with a coverage mechanism. To ensure a thorough assessment, multiple evaluation metrics are employed, including ROUGE-1, ROUGE-2, ROUGE-L, METEOR, and BERTScore. By analyzing the performance of these models, this study identifies their respective strengths and limitations in the summarization of medical text. The findings of this paper provide helpful information for medical professionals who need automated summarization solutions in the healthcare sector.