Fine-grained Abnormality Detection and Natural Language Description of Medical CT Images Using Large Language Models
Journal Title: International Journal of Innovative Research in Computer Science and Technology - Year 2025, Vol 13, Issue 1
Abstract
Medical report generation demands accurate abnormality detection and precise description generation from CT images. While large language models have shown promising results in natural language processing tasks, their application in medical imaging analysis faces challenges due to the complexity of fine-grained feature detection and the requirement for domain-specific knowledge. This paper presents a novel framework integrating large language models with specialized medical image processing techniques for fine-grained abnormality detection and natural language description generation. Our approach incorporates a multi-modal knowledge enhancement module and a hierarchical attention mechanism to bridge the gap between visual understanding and textual description. The framework employs an adapter-based architecture for efficient domain adaptation and introduces a medical knowledge-enhanced loss function to improve description accuracy. Experimental results on three public datasets demonstrate the effectiveness of our approach, achieving 94.6% detection accuracy and a BLEU-4 score of 0.421 for description generation, surpassing current state-of-the-art methods. The system shows particular strength in handling subtle abnormalities, with a 91.2% average precision in fine-grained detection tasks. Comprehensive ablation studies validate the contribution of each component, while qualitative analysis demonstrates the clinical relevance of generated descriptions. The proposed framework represents a significant advancement in automated medical image analysis, offering potential benefits for clinical workflow optimization and diagnostic support.
Authors and Affiliations
Zhongwen Zhou , Siwei Xia , Mengying Shu , Hong Zhou
Keywords
Related Articles
- EP ID EP753591
- DOI 10.55524/ijircst.2024.12.6.8
- Views 39
- Downloads 0
Using Big Data Technique for Building Edit Alert System for Wikipedia Infoboxes Based on MapReduce Method
Wikipedia is an online encyclopedia and has become a vital information resource for users as well as for many knowledge bases derived from it. This information requires manual editing for update. Wikipedia provides an in...
Verification of Adaptive Collection for Brain Computer Interface
To provide speech prostheses for individuals with severe communication impairments, brain computer interfaces (BCIs) using silent speech have been studied. I proposed adaptive collection, which divided brainwaves into sm...
A Review Study on Medicinal Properties of Psidium Guajava
Guava is a plant local to Tropical America and one of the most well-known in the Myrtaceae family. In contrast with different organic products, guava is untreated with synthetic compounds, making it a better choice. It h...
Approaches of Data Warehousing and Their Applications: A Review
A data warehouse, DW in short is a huge repository of corporate data that is employed to aid an organization's decision-making. The data warehouse idea has been around throughout eighties, while it was created to assist...
HSV Values and OpenCV for Object Tracking
This research shows how colour and motion may be utilised to speed up the surveillance of things. Video tracing is a technique for detecting a huge vehicle over a long distance using a camera. The main goal of video trac...