AUTHOR=He Quan-Hao , Tan Hao , Liao Fang-Tong , Zheng Yi-Neng , Lv Fa-Jin , Jiang Qing , Xiao Ming-Zhao TITLE=Stratification of malignant renal neoplasms from cystic renal lesions using deep learning and radiomics features based on a stacking ensemble CT machine learning algorithm JOURNAL=Frontiers in Oncology VOLUME=Volume 12 - 2022 YEAR=2022 URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2022.1028577 DOI=10.3389/fonc.2022.1028577 ISSN=2234-943X ABSTRACT=Using nephrographic phases CT images combined with pathology diagnosis, we aim to develop and validate a fusion feature-based stacking ensemble machine learning model to distinguish malignant renal neoplasms from cystic renal lesions (CRL). This retrospective research includes 166 individuals with CRL for model training and 47 individuals with CRL in another institution for model testing. Histopathology results are adopted as diagnosis criterion. Nephrographic phases CT scans are selected to build the fusion feature-based machine learning algorithms. Pretrained 3D-ResNet50 CNN model and radiomics methods are selected to extract deep features and radiomics features respectively.5-fold cross validated least absolute shrinkage and selection operator (LASSO) regression methods are adopted to identify the most discriminative candidate features in the development cohort. Intra-class correlation coefficients and inter-class correlation coefficients are employed to evaluate feature’s reproducibility. Pearson correlation coefficients for normal distribution features and Spearman’s rank correlation coefficients for non-normal distribution features are used to eliminate redundant features. After that, stacking ensemble machine learning models are developed in the training cohort. The area under the receiver operator characteristic curve (ROC), calibration curve and decision curve analysis (DCA) are adopted in the testing cohort to evaluate the performance of each model. The stacking ensemble machine learning algorithm reached excellent diagnostic performance in testing dataset. Calibration plot shows well stability when using stacking ensemble model. Net benefits presented by DCA are higher than Bosniak-2019 version classification when employing any machine learning algorithms. Fusion feature-based machine learning algorithm accurately distinguishes malignant renal neoplasms from CRL which outperformed the Bosniak-2019 version classification and demonstrates to be more applicable for clinical decision-making.