Nowadays, texture analysis on computed tomography images has broad perspectives in the differentiation of different pathological diagnosis of renal tumors, and computed tomography texture analysis (CTTA) can accurately distinguish benign or malignant tumors of the kidney. Retrospective case-control studies or prospective studies can be chosen for the research designs, and CTTA can be performed using open source databases and hospital computed tomography image databases of benign Tumors, such as angiomyolipoma, adenoma, fibroid, lipoma, myoma, hemangioma, or renal malignancies such as Wilms tumor, fibrosarcoma, myosarcoma, liposarcoma, papillary renal cell carcinoma, clear cell renal cell carcinoma, chromophobe renal cell carcinoma. Then we can choose the two categories of different pathological diagnosis as the target of classification (it is better to choose malignant and benign tumors for comparison). The reviewers first contour the region of interest from the cross-sectional computed tomography images, and then extract the texture feature from the region of interest. In other words, shape features, grey level histogram features, grey level co-occurrence matrix features and Gabor features are extracted. The reviewers use feature selection methods to remove redundant features for selection of the most accurate model for distinguishing different pathological diagnosis of renal tumors. The reviewers implement each feature selection model with different classifiers and different modes of validation in software to train and compare these settings with the parameters of prediction performances. In general, lesion uniformity is low and lesion entropy is high in malignant tumors. If a model combining several texture features can obtain an area under the receiver operating curve more than >0.8, it is feasible to use CTTA to distinguish different pathological diagnosis of renal tumors. It needs to be further verified in different ethnic groups before it can be clinically applied and further generalized to different populations.