End-to-end prostate lesion segmentation and detection from undersampled diffusion-weighted MRI
Abstract
Automated prostate lesion detection from diffusion-weighted MR imaging (DWI) conventionally relies on reconstructed images and does not fully use all acquisition information. We present an end-to-end network that uses complex multicoil DWI data and coil sensitivity information for prostate lesion segmentation. We included 269 fastMRI prostate examinations with additional voxel-level lesion annotations. Of these, 64 examinations contained PIRADS 4 or 5 lesions. The proposed network, VarNet-Seg, consists of a k-space variational network (VarNet) with six variational cascades and an image-space 3D segmentation U-Net. Comparator models used the same segmentation architecture with fully coil-combined images (CC-Seg), fixed root-sum-of-squares reconstruction (RSS-Seg), or fixed SENSE reconstruction (SENSE-Seg). We performed ablations with end-to-end training and data consistency (DC) enabled or disabled. We evaluated lesion segmentation using Dice and patient-level detection using AUROC in five-fold cross-validation with full sampling, simulated undersampling, added noise, or both. VarNet-Seg achieved the highest Dice at every undersampling level. At eight-fold acceleration, Dice was 0.22 for VarNet-Seg, 0.14 for SENSE-Seg and 0.16 for VarNet-Seg (no DC). The paired difference from SENSE-Seg was 0.08 (95% CI, 0.04-0.11). With full sampling, VarNet-Seg and RSS-Seg performed similarly (0.40 vs 0.39). Patient-level AUROC decreased from 0.75 with full sampling to 0.63 at eight-fold acceleration; corresponding RSS-Seg AUROCs were 0.76 and 0.64. The proposed multicoil network improved prostate lesion segmentation under simulated undersampling compared to conventional models. Matched ablations supported contributions from both end-to-end training and data consistency. Prospective evaluation is needed to establish whether this approach can be used clinically in accelerated prostate DWI.
Related
- Resolved principal component analysis: A resolved point spread function reconstruction method for undersampled images
- An interim analysis of acute treatment adverse events and patient reported quality of life for functional imaging guided salivary gland sparing radiation therapy using diffusion-weighted magnetic resonance imaging.
- Benchmarking and sensitivity analysis of segmentation methods for image-based fiber detection in composites
- Improving the Sensitivity of Gravitational Wave Detection with Weighted Conformal Prediction
- A novel automatic real-time motion tracking method in MRI-guided radiotherapy using enhanced Tracking-Learning-Detection framework with automatic segmentation.
BibTeX
@misc{rotkopf2026endtoend,
title = {{End-to-end prostate lesion segmentation and detection from undersampled diffusion-weighted MRI}},
author = {Lukas T. Rotkopf and Marco Schlimbach and Kevin Kröninger and Stefan O. Schoenberg and Heinz-Peter Schlemmer and Jens Kleesiek and Moritz Rempe},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.33554},
eprint = {2610.33554},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.33554},
}
RIS
TY - UNPB
TI - End-to-end prostate lesion segmentation and detection from undersampled diffusion-weighted MRI
AU - Lukas T. Rotkopf
AU - Marco Schlimbach
AU - Kevin Kröninger
AU - Stefan O. Schoenberg
AU - Heinz-Peter Schlemmer
AU - Jens Kleesiek
AU - Moritz Rempe
PY - 2026
DA - 2026/10/24
PB - arXiv
DO - 10.48550/arxiv.2610.33554
UR - https://arxiv.org/abs/2610.33554
AB - Automated prostate lesion detection from diffusion-weighted MR imaging (DWI) conventionally relies on reconstructed images and does not fully use all acquisition information. We present an end-to-end network that uses complex multicoil DWI data and coil sensitivity information for prostate lesion segmentation. We included 269 fastMRI prostate examinations with additional voxel-level lesion annotations. Of these, 64 examinations contained PIRADS 4 or 5 lesions. The proposed network, VarNet-Seg, consists of a k-space variational network (VarNet) with six variational cascades and an image-space 3D segmentation U-Net. Comparator models used the same segmentation architecture with fully coil-combined images (CC-Seg), fixed root-sum-of-squares reconstruction (RSS-Seg), or fixed SENSE reconstruction (SENSE-Seg). We performed ablations with end-to-end training and data consistency (DC) enabled or disabled. We evaluated lesion segmentation using Dice and patient-level detection using AUROC in five-fold cross-validation with full sampling, simulated undersampling, added noise, or both. VarNet-Seg achieved the highest Dice at every undersampling level. At eight-fold acceleration, Dice was 0.22 for VarNet-Seg, 0.14 for SENSE-Seg and 0.16 for VarNet-Seg (no DC). The paired difference from SENSE-Seg was 0.08 (95% CI, 0.04-0.11). With full sampling, VarNet-Seg and RSS-Seg performed similarly (0.40 vs 0.39). Patient-level AUROC decreased from 0.75 with full sampling to 0.63 at eight-fold acceleration; corresponding RSS-Seg AUROCs were 0.76 and 0.64. The proposed multicoil network improved prostate lesion segmentation under simulated undersampling compared to conventional models. Matched ablations supported contributions from both end-to-end training and data consistency. Prospective evaluation is needed to establish whether this approach can be used clinically in accelerated prostate DWI.
ER -