Deep Neural Patchworks Predict Renal Imaging Biomarkers From Non-contrast MRI via Knowledge Transfer From Arterial-Phase Contrast-Enhanced MRI.
Authors
Affiliations (5)
Affiliations (5)
- Department of Diagnostic and Interventional Radiology, Medical Center - University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany (K.F.K., A.F., S.R., F.B., M.F.R.). Electronic address: [email protected].
- Department of Diagnostic and Interventional Radiology, Medical Center - University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany (K.F.K., A.F., S.R., F.B., M.F.R.).
- Medical Physics, Department of Radiology, Medical Center - University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany (M.R., E.K., J.M.N.).
- Medical Physics, Department of Radiology, Medical Center - University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany (M.R., E.K., J.M.N.); Department of Nephrology, Medical Center - University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany (J.M.N.).
- Institute of Epidemiology and Prevention, Medical Center - University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany (A.K., P.S.).
Abstract
Contrast-enhanced (CE) MRI provides clear corticomedullary contrast for renal compartment delineation but may be contraindicated or undesirable in routine practice. We aimed to enable automated extraction of renal imaging biomarkers from routine non-contrast-enhanced (NCE) T1-weighted MRI by transferring CE-derived compartment labels. This retrospective single-center study (January 2017 to December 2021) included 200 patients with paired arterial-phase CE and NCE T1-weighted MRI. Cortex, medulla, and sinus were manually segmented on CE MRI and rigidly transferred to NCE MRI to provide voxel-level reference labels. A hierarchical 3D Deep Neural Patchworks (DNP) model was trained on 100 examinations (90 training/10 validation) and evaluated on an independent test set of 100 examinations using the transferred CE masks on NCE as reference. Performance was assessed using Dice similarity of segmentations and biomarker agreement using volumes (Pearson, MAE, Lin's CCC, and Bland-Altman). Whole-kidney segmentation Dice was 0.950 (left) and 0.953 (right). Total kidney volume showed high agreement with minimal bias (MAE 8.76 mL, 2.5% of mean; CCC 0.983; bias -1.56mL; 95% limits of agreement -28.81 to 25.69 mL). Cortex volume was modestly overestimated and medulla volume underestimated, shifting predicted compartment fractions toward cortex (74.7% vs. 72,1% in ground truth; medulla 21.5% vs. 24.3%; sinus 3.8% vs. 3.6%. Sinus volume maintained high concordance despite higher Dice dispersion. CE-supervised knowledge transfer enables accurate, well-calibrated total kidney volumetry from routine NCE MRI and supports contrast-free renal biomarker extraction in kidneys without major structural distortion.