Benchmarking multiorgan segmentation tools for multiparametric T1-weighted abdominal MRI
Year: 2025
Type: conference-paper
Abstract: The segmentation of multiple organs in multi-parametric MRI studies is critical for many applications in radiology, such as correlating imaging biomarkers with disease status (e.g., cirrhosis, diabetes). Recently, three publicly available tools, such as MRSegmentator (MRSeg), TotalSegmentator MRI (TS), and TotalVibeSegmentator (VIBE), have been proposed for multi-organ segmentation in MRI. However, the performance of these tools on specific MRI sequence types has not yet been quantified. In this work, a subset of 40 volumes from the public Duke Liver Dataset was curated. The curated dataset contained 10 volumes each from the pre-contrast fat saturated T1, arterial T1w, venous T1w, and delayed T1w phases, respectively. Ten abdominal structures were manually annotated in these volumes. Next, the performance of the three public tools was benchmarked on this curated dataset. The results indicated that MRSeg obtained a Dice score of 80.7 ± 18.6 and Hausdorff Distance (HD) error of 8.9 ± 10.4 mm. It fared the best (p < .05) across the different sequence types in contrast to TS and VIBE.
DOI: 10.1117/12.3048938
Authors: Nicole Tran, Anisa V. Prasad, Yan Zhuang, Tejas Sudharshan Mathai, Boah Kim, Sydney V. Lewis, Pritam Mukherjee, Jianfei Liu, Ronald M. Summers
Institution: National Institutes of Health
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4409158800