Explainable Machine-learning Model Based on Multimodal Ultrasound for Non-invasive Detection of Early Renal Fibrosis: A Multicenter Study.
Authors
Affiliations (17)
Affiliations (17)
- Department of Ultrasound, Renmin Hospital of Wuhan University, Wuhan, Hubei, China (Y.Z., X.H., Q.D., Q.Z.).
- Department of Ultrasound, The First People's Hospital of Jingzhou, Jingzhou, Hubei, China (W.X., W.L.).
- Department of Ultrasound, Yichang Central People's Hospital, Yichang, Hubei, China (C.Z., D.S.).
- Department of Ultrasound, Ezhou Central Hospital, Ezhou, Hubei, China (W.Y., Y.J.).
- Department of Ultrasound, Huanggang Central Hospital, Huanggang, Hubei, China (X.Z., R.L.).
- Department of Ultrasound, The Central Hospital of Enshi Tujia and Miao Autonomous Prefecture, Enshi, Hubei, China (Y.H., Y.X.).
- Department of Ultrasound, Xiaogan Central Hospital, Xiaogan, Hubei, China (Y.T., X.D.).
- Department of Ultrasound, China Resources & Wisco General Hospital, Wuhan, Hubei, China (H.Y., B.L.).
- Department of Ultrasound, Xiangyang No.1 People' s Hospital, Xiangyang, Hubei, China (L.G., J.Y.).
- Department of Ultrasound, Affiliated Hospital of Hubei University of Medicine, Shiyan, Hubei, China (W.Z., X.G.).
- Department of Ultrasound, Huangshi Central Hospital, Huangshi, Hubei, China (Z.H., X.R.).
- Department of Ultrasound, Wuhan Third Hospital, Wuhan, Hubei, China (Y.P., C.Z.).
- Department of Ultrasound, Macheng People' s Hospital, Macheng, Hubei, China (Y.W., X.W.).
- Department of Ultrasound, The Fifth Hospital of Wuhan, Wuhan, Hubei, China (J.M., J.W.).
- Department of Ultrasound, Hubei Provincial Hospital of Traditional Chinese Medicine, Wuhan, Hubei, China (W.Y., L.X.).
- Department of Ultrasound, Xiantao No.1 People's Hospital, Xiantao, Hubei, China (X.T., S.C.).
- Department of Ultrasound, Renmin Hospital of Wuhan University, Wuhan, Hubei, China (Y.Z., X.H., Q.D., Q.Z.). Electronic address: [email protected].
Abstract
To develop and validate an interpretable, multimodal, ultrasound-based machine-learning (ML) model for the non-invasive identification of early renal fibrosis in patients with chronic kidney disease (CKD). In this prospective, multicenter study, 369 participants (healthy controls, n = 161; mild fibrosis, n = 208) were recruited from 16 institutions. Mild fibrosis was defined by histopathological grading. Six ML algorithms were evaluated, and multiple models were constructed using different combinations of conventional ultrasound, ultra micro angiography (UMA), shear-wave elastography (SWE), super-resolution ultrasound (SRUS), and clinical variables; performance was compared in terms of discrimination, calibration, and clinical utility. External validation was conducted using held-out centers. Model interpretability was examined using Shapley Additive exPlanations (SHAP). In the internal dataset, the comprehensive fusion model integrating SRUS, SWE, conventional ultrasound, and clinical variables, based on Light Gradient Boosting Machine, achieved an area under the curve (AUC) of 0.948 (accuracy, 0.880; sensitivity, 0.987). In the external dataset, it maintained good generalization (AUC, 0.823; accuracy, 0.748; sensitivity, 0.883). The fusion model outperformed the clinical baseline model in both internal and external validations (internal AUC, 0.732; external AUC, 0.650). SHAP analysis identified both imaging parameters (vessel density, fractal dimension) and clinical indices (serum creatinine (Scr), estimated glomerular filtration rate (eGFR)) as key predictors, consistent with pathophysiological changes of early fibrosis. An explainable multimodal ultrasound-based ML model showed promising performance for the non-invasive identification of early renal fibrotic changes and may support adjunctive risk stratification in CKD.