August 9, 2025

Improving Radiology Report Generation with Semantic Understanding.

Puntos clave

The proposed RRG-LLM model improves radiology report generation by utilizing semantic understanding and minimal resources.
Finetuning using LoRA allows the model to adapt effectively to the medical domain for better output.
Performance metrics showed a 51.7% improvement in ROUGE-L and a 42.85% increase in METEOR scores compared to the baseline.
This advancement in report generation could significantly facilitate accurate communication in healthcare settings.

Resumen

This study proposes RRG-LLM, a model designed to enhance RRG by effectively learning medical domain with minimal computational resources. Initially, LLM is finetuned by LoRA, enabling efficient adaptation to the medical domain. Subsequently, only the linear projection layer that project the image into text is finetuned to extract important information from the radiology image and project it onto the text dimension. Proposed model demonstrated notable improvements in report generation. The performance of ROUGE-L was improved by 0.096 (51.7%) and METEOR by 0.046 (42.85%) compared to the baseline model.

Me gusta

Guardar