Automated Vertebral Heart Size Estimation from Thoracic Radiographs in Dogs with AI-assisted Clinical Decision Support.

AI-assisted VHS automation enhances cardiomegaly detection accuracy and workflow efficiency in canine radiology.

This study presents an AI-assisted framework for automated Vertebral Heart Size (VHS) estimation from canine thoracic radiographs, addressing the clinical need for rapid and accurate cardiomegaly detection. The integrated system combines deep learning-based computer vision to detect thoracic anatomical landmarks and calculate VHS measurements with a Large Language Model that generates structured clinical summaries. The framework processes DICOM radiographs, extracts relevant metadata, and produces preliminary clinical assessments within seconds of image upload. Quantitative evaluation demonstrates accurate landmark detection and reliable VHS computation, while qualitative assessment confirms that generated summaries are coherent, contextually appropriate, and consistent with radiographic findings. This multimodal AI approach substantially reduces time required for manual measurement and report generation while minimizing inter-observer variability. The results indicate that such systems have significant potential to enhance veterinary radiology workflows, improve clinical decision-making efficiency, and enable timely detection of cardiac enlargement in companion animals, ultimately supporting better treatment planning for dogs with cardiac disease.

This summary was distilled by AI and may occasionally misinterpret data. Confirm critical details with the primary literature before clinical application.