Survival analysis was performed to predict the OS and PFS; then, binary classification models were created at different endpoints. For the classification models, 7 different classifiers were trained: random forest (RFT), gradient boosting, Adaboost, logistic regression, K-neighbours, naïve Bayes, and SVM. All the algorithms were taken from the Scikit-learn Python library. Survival analysis differed from classical regression by the existence of censored data. Survival random forest [29,30] is a modified random forest algorithm that can perform such an analysis by calculating the survival function. This function gives the probability that one patient survives longer than a specific time. It can also calculate the risk score, which is a value computed on an arbitrary scale. Then, if samples are ordered according to their predicted risk score (in ascending order), one obtains the sequence in time of events, as predicted by the model. We used the RandomSurvivalForest from the Scikit-survival Python library to perform such a regression.

Note: The content above has been extracted from a research article, so it may not display correctly.



Q&A
Please log in to submit your questions online.
Your question will be posted on the Bio-101 website. We will send your questions to the authors of this protocol and Bio-protocol community members who are experienced with this method. you will be informed using the email address associated with your Bio-protocol account.



We use cookies on this site to enhance your user experience. By using our website, you are agreeing to allow the storage of cookies on your computer.