XCAT-2.0: A Comprehensive Library of Personalized Digital Twins Derived from CT Scans

• DOI: 10.48550/arxiv.2405.11133
• Publication Date: 2024-05-17
• Authors: Lavsen Dahal; Mobina Ghojoghnejad; Dhrubajyoti Ghosh; Yubraj Bhandari; David Kim; Fong Chi Ho; Fakrul Islam Tushar; Ehsan Abadi; Ehsan Samei; Joseph Y. Lo; Paul Segars
• Keywords: FOS: Computer and information sciences; Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV); Computer Science - Computer Vision and Pattern Recognition; FOS: Electrical engineering, electronic engineering, information engineering; Electrical Engineering and Systems Science - Image and Video Processing

Virtual Imaging Trials (VIT) offer a cost-effective and scalable approach for evaluating medical imaging technologies. Computational phantoms, which mimic real patient anatomy physiology, play central role in VIT. However, the current libraries of computational phantoms face limitations, particularly terms sample size diversity. Insufficient representation population hampers accurate assessment technologies across different groups. Traditionally, were created by manual segmentation, is laborious time-consuming task, impeding expansion phantom libraries. This study presents framework realistic modeling using suite four deep learning segmentation models, followed three forms automated organ quality control. Over 2500 with up to 140 structures illustrating sophisticated detailed anatomical are released. Phantoms available both voxelized surface mesh formats. The aggregated an in-house CT scanner simulator produce images. can potentially advance virtual trials, facilitating comprehensive reliable evaluations may be requested at https://cvit.duke.edu/resources/, code, model weights, images https://xcat-2.github.io.