AUTHOR=Chen Wenbo , Zhang Lu , Cai Guanhui , Zhang Bin , Lian Zhouyang , Li Jing , Wang Wenjian , Zhang Yuxian , Mo Xiaokai 

TITLE=Machine learning-based multimodal MRI texture analysis for assessing renal function and fibrosis in diabetic nephropathy: a retrospective study

JOURNAL=Frontiers in Endocrinology

VOLUME=Volume 14 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/endocrinology/articles/10.3389/fendo.2023.1050078

DOI=10.3389/fendo.2023.1050078

ISSN=1664-2392

ABSTRACT=Abstract     
To determine the possible practicability of Machine Learning(ML)-based Multimodal MRI texture analysis (mMRI-TA) for assessing renal function and fibrosis in diabetic nephropathy (DN). For this retrospective study, 71 patients (between 1 January 2013 and 1 January 2021) were included and randomly assigned to the training cohort (n1=50) and the testing cohort (n2=21). According to estimated glomerular filtration rate (eGFR), patients were grouped into normal renal function (normal-RF) group; non-severe renal function impairment (non-sRI) group and severe renal function impairment (sRI) group. Based on the largest coronal image of T2WI, the Speeded up robust features (SURF) algorithm was used for texture features extraction. Analysis of Variance (ANOVA), Relief and Recursive Feature Elimination (RFE) were applied to select the important features and then support vector machine (SVM), logistic regression (LR) and Random Forest (RF) were used for the model construction. The values of area under the curve (AUC) on receiver operating characteristic (ROC) curve analysis were used to assess their performance. The robust T2WI model was selected to construct a multimodal MRI model by combining the measured values on BOLD (Blood oxygenation level-dependent) and diffusion-weighted (DW) imaging. The mMRI-TA model achieved robust and excellent performance in classifying the sRI group, non-sRI group and normal-RF group, with an AUC of 0.978 (95% confidence interval [CI]:0.963-0.993), 0.852 (95% CI: 0.798-0.902 ), 0.972 (95% CI: 0.995-1.0) respectively in the training cohort and 0.961 (95% CI: 0.8529-1.0), 0.809(95% CI: 0.60-0.980) and 0.850 (95% CI: 0.6375-0.988) respectively in the testing cohort.  The model built from multimodal MRI in diabetic nephropathy (DN) outperformed in assessing renal function and fibrosis. Compared to the single T2WI sequence, mMRI-TA can improve the performance in assessing renal function.