AUTHOR=Ma Feier , Shao Xian , Zhang Yuling , Li Jinlao , Li Qiuhong , Sun Haizhen , Wang Tongdan , Liu Hongyan , Zhao Feiyu , Chen Lianqin , Chen Jiamian , Zhou Saijun , Ji Qian , Yu Pei TITLE=An arterial spin labeling−based radiomics signature and machine learning for the prediction and detection of various stages of kidney damage due to diabetes JOURNAL=Frontiers in Endocrinology VOLUME=Volume 15 - 2024 YEAR=2024 URL=https://www.frontiersin.org/journals/endocrinology/articles/10.3389/fendo.2024.1333881 DOI=10.3389/fendo.2024.1333881 ISSN=1664-2392 ABSTRACT=Objective The aim of this study is to assess the predictive capabilities of a radiomics signature obtained from arterial spin labeling (ASL) imaging in forecasting and detecting stages of kidney damage in patients with diabetes mellitus (DM), as well as to analyze the correlation between texture feature parameters and clinical biological indicators. Additionally, this study seeks to identify imaging risk factors associated with early renal injury in diabetic patients, with the ultimate goal of offering novel insights for predicting and diagnosing early renal injury and its progression in DM patients.  Materials and methods Forty-two healthy volunteers (Group A), sixty-eight individuals with diabetes (Group B) exhibiting microalbuminuria, and fifty-three patients with diabetic nephropathy (Group C) were included in the study. ASL using magnetic resonance imaging (MRI) at 3.0T was conducted. The radiologist manually delineated regions of interest (ROI) on the ASL maps of both the right and left kidney cortex. Texture features from the ROI were extracted utilizing MaZda software. The radiologist manually delineated regions of interest (ROI) on the ASL maps of both the right and left kidney cortex. Texture features from the ROI were extracted utilizing MaZda software. Results A total of 367 texture features were extracted from the ROI in the kidney and refined based on selection criteria using MaZda software across groups A, B, and C. The renal blood flow (RBF) values of the renal cortex in groups A, B, and C exhibited a decreasing trend, with values of 256.458±54.256 mL/100g/min,213.846±52.109 mL/100g/min, and 170.204±34.992 mL/100g/min, respectively.Additionally, a comprehensive prediction model combining imaging labels and biological indicators, with the naive Bayes machine learning algorithm as the best model, demonstrated an AUC of 0.734, accuracy of 0.74, and precision of 0.43.  Conclusion ASL imaging sequences have demonstrated the ability to accurately detect alterations in kidney function and blood flow in patients with DM.