This is the authors' abstract. We don't add key points for this paper.
Background: The increasing prevalence of colorectal cancer (CRC) in the past three decades in Iran has made it as a key public health burden. The goal of the study was to predict the metastasis in CRC patients using machine learning approach. Methods: This study is focused on 1127 CRC patients, who underwent proper treatments in tertiary Taleghani Hospital. The machine learning approach including Naive Bayes (NB), Random Forest (RF), Support Vector Machine (SVM), Neural Network (NN), Regression Tree (RT) and Logistic Regression (LR) were used for the prediction of metastasis in CRC. Receiver operating characteristic (ROC) curve, sensitivity, specificity and AUC were carried out to evaluate the performance of the approaches Results: Out of 1127 patients, 183(16%) experienced metastasis. In prediction of metastasis, the NN and RF had the most sensitivity. Also the specificity of NN and DT were greater in comparison to other methods. The NN had the highest AUC (0.99%), topmost sensitivity (98.5%) and accuracy (99.6%).Conclusion: Both tumor stage and size of tumor were regarded as indispensable variables in RF model, whereas the NN generally outperformed other ML-based approaches.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Borumandnia et al. (2022) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: