Dynamic parametric MRI and deep learning: Unveiling renal pathophysiology through accurate kidney size quantification

• Published in: NMR in biomedicine Vol. 37; no. 4; pp. e5075 - n/a
• Main Authors: Klein, Tobias, Gladytz, Thomas, Millward, Jason M., Cantow, Kathleen, Hummel, Luis, Seeliger, Erdmann, Waiczies, Sonia, Lippert, Christoph, Niendorf, Thoralf
• Format: Journal Article
• Language: English
• Published: England Wiley Subscription Services, Inc 01.04.2024
• Subjects: Algorithms; Animals; Aorta; Biomarkers; Contrast agents; Contrast media; Deep Learning; Furosemide; Hypoxemia; Image Processing, Computer-Assisted; kidney; Kidney - diagnostic imaging; Kidney diseases; kidney size; Kidneys; Machine learning; Magnetic Resonance Imaging; MRI; Occlusion; Organophosphorus Compounds; parametric mapping; Pathophysiology; Performance measurement; Rats; Segmentation; Triazoles; deep learning; kidney; parametric mapping; segmentation; MRI; kidney size
• ISSN: 0952-3480; 1099-1492; 1099-1492
• DOI: 10.1002/nbm.5075

Renal pathologies often manifest as alterations in kidney size, providing a valuable avenue for employing dynamic parametric MRI as a means to derive kidney size measurements for the diagnosis, treatment, and monitoring of renal disease. Furthermore, this approach holds significant potential in supporting MRI data‐driven preclinical investigations into the intricate mechanisms underlying renal pathophysiology. The integration of deep learning algorithms is crucial in achieving rapid and precise segmentation of the kidney from temporally resolved parametric MRI, facilitating the use of kidney size as a meaningful (pre)clinical biomarker for renal disease. To explore this potential, we employed dynamic parametric T2 mapping of the kidney in rats in conjunction with a custom‐tailored deep dilated U‐Net (DDU‐Net) architecture. The architecture was trained, validated, and tested on manually segmented ground truth kidney data, with benchmarking against an analytical segmentation model and a self‐configuring no new U‐Net. Subsequently, we applied our approach to in vivo longitudinal MRI data, incorporating interventions that emulate clinically relevant scenarios in rats. Our approach achieved high performance metrics, including a Dice coefficient of 0.98, coefficient of determination of 0.92, and a mean absolute percentage error of 1.1% compared with ground truth. The DDU‐Net enabled automated and accurate quantification of acute changes in kidney size, such as aortic occlusion (−8% ± 1%), venous occlusion (5% ± 1%), furosemide administration (2% ± 1%), hypoxemia (−2% ± 1%), and contrast agent‐induced acute kidney injury (11% ± 1%). This approach can potentially be instrumental for the development of dynamic parametric MRI‐based tools for kidney disorders, offering unparalleled insights into renal pathophysiology. Renal pathologies can be detected through alterations in kidney size, making dynamic parametric MRI a valuable tool for deriving kidney size measurements in the diagnosis, treatment, and monitoring of renal disease. This study utilized dynamic parametric T2 mapping and a custom‐tailored deep dilated U‐Net architecture, achieving exceptional performance in accurately quantifying acute changes in kidney size. This innovative approach has significant potential for developing MRI‐based tools and advancing our understanding of renal pathophysiology.