Workflow for automatic renal perfusion quantification using ASL‐MRI and machine learning

Abstract

This is an open access article under the terms of the Creat ive Commo ns Attri bution-NonCo mmercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes.

Optimization of Fair Arterial Spin Labeling Magnetic Resonance Imaging (ASL-MRI) for Renal Perfusion Quantification in Dogs: Pilot Study

“…There is a lot of variation in the literature when it comes to the values of λ that are used to quantify RBF with ASL-MRI. In humans, values ranging from 0.80 to 0.94 mL/g have been applied [ 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 ]. In pigs, the λ value even varied between 0.80 and 1 mL/g [ 48 , 49 , 50 , 51 ].…”

“…In pigs, the λ value even varied between 0.80 and 1 mL/g [ 48 , 49 , 50 , 51 ]. In many human studies [ 23 , 24 , 25 , 26 ] and even in animal studies (rabbits and mice) [ 52 , 53 , 54 ], the mean λ value of human brain tissue (0.9 mL/g) was used as an equivalent for kidney λ because of the assumption that these λ values are similar [ 28 ]. However, no comprehensive studies have been conducted to confirm this statement or clarify whether it is applicable to all species of animals.…”

“…In the literature, diverse values of λ are used to quantify renal perfusion using ASL-MRI in both humans and rodents. [ 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 ]. The brain λ-value is often used as it is assumed to be similar to the kidney λ-value [ 28 ].…”

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

“…[58][59][60] Most of these deep learning (DL) approaches take advantage of U-Net variants. [61][62][63][64][65] Recent application of DL-based kidney segmentation has focused on automation of renal cyst and kidney volume measurements in healthy subjects and patients with autosomal-dominant PKD and CKD. 54,56,60,[66][67][68][69][70] The feasibility and reliability of dynamic or longitudinal MRIbased KS monitoring using DL in acute pathophysiological scenarios, where changes may be more subtle than in autosomal-dominant PKD or CKD, has not yet been investigated.…”

“…MRI studies using neural networks for renal segmentation reported processing times as good as 1–10 s per subject 58–60 . Most of these deep learning (DL) approaches take advantage of U‐Net variants 61–65 . Recent application of DL‐based kidney segmentation has focused on automation of renal cyst and kidney volume measurements in healthy subjects and patients with autosomal‐dominant PKD and CKD 54,56,60,66–70 .…”