3D Adversarial Segmentation of Kidney-Transplant Across Multiple MRI Sequences Using Probabilistic and Anatomical Priors
| dc.contributor.author | Sharaby, Israa | |
| dc.contributor.author | Balaha, Hossam Magdy | |
| dc.contributor.author | Ghazal, Mohammed | |
| dc.contributor.author | ETAL.. | |
| dc.date.accessioned | 2026-06-30T08:58:12Z | |
| dc.date.available | 2026-06-30T08:58:12Z | |
| dc.date.issued | 2026 | |
| dc.description | Kidney transplantation is the definitive treatment for end-stage kidney disease, offering better outcomes and quality of life than prolonged dialysis [1]. However, the post-transplant period remains vulnerable to both acute rejection in the early phase and chronic allograft injury driven by fibrotic processes in the long term, making early and reliable assessment essential for preserving graft function [2,3]. Automated kidney segmentation is an essential prerequisite for quantitative kidney measurements because inaccurate delineation may propagate errors into downstream biomarker estimation and confound clinical interpretation [4]. Automated kidney segmentation in transplant magnetic resonance imaging (MRI), however, remains an open problem that presents different challenges from those encountered in native kidney segmentation, as shown in Figure 1. Surgical graft placement introduces substantial variability in kidney orientation, position, and morphology across patients, while MRI exhibits heterogeneous intensity distributions, noise, and low contrast between renal parenchyma and surrounding tissues, resulting in ambiguous boundaries that are difficult to delineate reliably. | |
| dc.description.abstract | Background/Objectives: Accurate kidney segmentation from magnetic resonance imaging (MRI) in kidney-transplant patients is essential for quantitative graft assessment, yet it remains challenging due to low tissue contrast, intensity inhomogeneity, and inter-patient anatomical variability introduced by surgical graft placement. Methods: We propose a 3D adversarial segmentation framework that incorporates probabilistic appearance and anatomical shape priors into a residual conditional generative adversarial network (GAN). The framework integrates image-driven and prior-guided information to improve boundary delineation under challenging imaging conditions and is evaluated on 100 kidney-transplant patients across T2-weighted imaging, BOLD-MRI, and DW-MRI using leave-one-out cross-validation. Results: The proposed method achieves mean Dice scores of 90.86% on T2-weighted imaging, 92.02% on BOLD-MRI, and 94.00% on DW-MRI. Consistent performance across all modalities demonstrates robustness under heterogeneous MRI acquisitions. The incorporation of prior guidance improves segmentation stability and anatomical consistency, particularly in low-contrast modalities. Conclusions: The proposed framework enables reliable kidney delineation across multiple MRI sequences, supporting consistent extraction of quantitative imaging biomarkers. This capability facilitates noninvasive assessment of renal graft function and supports longitudinal monitoring of transplant patients. keywords: anatomical prior, BOLD-MRI, DW-MRI, generative adversarial network, kidney-transplant, MRI segmentation, probabilistic appearance prior. | |
| dc.identifier.doi | https://doi.org/10.3390/diagnostics16091369 | |
| dc.identifier.uri | https://repository.adu.ac.ae/handle/1/8318 | |
| dc.language.iso | en | |
| dc.title | 3D Adversarial Segmentation of Kidney-Transplant Across Multiple MRI Sequences Using Probabilistic and Anatomical Priors | |
| dc.type | Article |