Anaconda3 platform

Manual tumor segmentation and feature extraction were both performed with the Radcloud platform (https://mics.radcloud.cn, Huiying Medical Technology Co., Ltd). Feature standardization, selection, and statistical analysis were performed with the Anaconda3 platform (https://www.anaconda.com/) using the Python 3.6 programming language (https://www.python.org/), mainly with the packages ‘scikit-learn’ (https://scikit-learn.org/) and ‘matplotlib’ (http://matplotlib.org/).

Although some of the radiomics features with an ICC lower than 0.75 were removed, there still remained a great quantity of features. In order to decrease the high degree of redundancy and irrelevance, feature selection was conducted using the least absolute shrinkage and selection operator (LASSO) regression method in Anaconda3 platform (https://www.anaconda.com) with scikit-learn (https://scikit-learn.org/) and matplotlib packages (https://matplotlib.org/).
The LASSO regression method has been proved to be efficient and effective in the high-dimensional data analysis [23 (link), 24 (link)]. It is aimed at minimizing the cost function and at keeping the features with nonzero coefficients. In this study, features passing the ICC screening were normalized by Z-score transform. Then, a 10-fold crossvalidation was carried out to choose the optimal parameters via the minimum of average mean square error. At last, the radiomics features with nonzero coefficients were used for further analysis.