Abstract
Chronic kidney disease (CKD) raises major concerns for global public health as it is characterized by high prevalence, low awareness, high healthcare costs, and poor prognosis. Therefore, our study prospectively established and validated native T(1) mapping-based radiomics models for the prediction of renal fibrosis and renal function in patients with CKD. Moreover, the area under the receiver operating characteristic curve (AUC) and diagnostic sensitivity, specificity, accuracy, positive predictive value, and negative predictive value were used to evaluate its performance. Thus, our results show that radiomics based on native T(1) mapping images can better identify renal function and renal fibrosis in patients with CKD and outperform conventional T(1) mapping parameters of ΔT(1) and T(1)%, thus providing more information for CKD management and clinical decision-making.