LegSegNet: A Public Deep Learning System for Lower Extremity CT Tissue Segmentation and Quantification
Researchers have developed LegSegNet, a novel deep learning system designed for segmenting and quantifying tissues in lower extremity CT scans. This system addresses limitations in existing tools by providing an end-to-end workflow for analyzing body composition, sarcopenia, and musculoskeletal diseases. LegSegNet achieves high accuracy, with an average Dice score of 89.31 on test data, and is the first publicly available system of its kind. AI
IMPACT Provides a new tool for medical image analysis, potentially accelerating research in body composition and disease monitoring.