Benchmarking Multi-Organ Segmentation Tools for Multi-Parametric T1-weighted Abdominal MRI

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Tran, Nicole, Prasad, Anisa, Zhuang, Yan, Mathai, Tejas Sudharshan, Kim, Boah, Lewis, Sydney, Mukherjee, Pritam, Liu, Jianfei, Summers, Ronald M.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913787497414656
author Tran, Nicole
Prasad, Anisa
Zhuang, Yan
Mathai, Tejas Sudharshan
Kim, Boah
Lewis, Sydney
Mukherjee, Pritam
Liu, Jianfei
Summers, Ronald M.
author_facet Tran, Nicole
Prasad, Anisa
Zhuang, Yan
Mathai, Tejas Sudharshan
Kim, Boah
Lewis, Sydney
Mukherjee, Pritam
Liu, Jianfei
Summers, Ronald M.
contents 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 $\pm$ 18.6 and Hausdorff Distance (HD) error of 8.9 $\pm$ 10.4 mm. It fared the best ($p < .05$) across the different sequence types in contrast to TS and VIBE.
format Preprint
id arxiv_https___arxiv_org_abs_2504_07729
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Benchmarking Multi-Organ Segmentation Tools for Multi-Parametric T1-weighted Abdominal MRI
Tran, Nicole
Prasad, Anisa
Zhuang, Yan
Mathai, Tejas Sudharshan
Kim, Boah
Lewis, Sydney
Mukherjee, Pritam
Liu, Jianfei
Summers, Ronald M.
Computer Vision and Pattern Recognition
Artificial Intelligence
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 $\pm$ 18.6 and Hausdorff Distance (HD) error of 8.9 $\pm$ 10.4 mm. It fared the best ($p < .05$) across the different sequence types in contrast to TS and VIBE.
title Benchmarking Multi-Organ Segmentation Tools for Multi-Parametric T1-weighted Abdominal MRI
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2504.07729