Artificial Intelligence-Based Prediction of Survival Outcomes Among Men with Prostate Cancer in Nigeria Using PSA Markers

One of the popularly recognised health problems among the Nigerian black men is metastatic prostate cancer, it is known to have variations in treatment response as well as the survival outcomes at the onset of surgical castration. Artificial intelligence and machine learning models have been classified to provide great opportunities for improving the assessment of the prognostics and individualised planning. Therefore, the study focused on a systematic review for the evaluation of its applications in terms of prognostic models used to predict the treatment outcomes at the onset of surgical castrations among the black men in Nigeria having the disease. The focus was on a systematic review using the PubMed, Google Scholar, Scopus based on clearly defined keywords. Studies that focused on the applications of machine learning or artificial intelligence for predicting survival outcome, treatment responses, progression of the disease as well as risk stratification after surgical castration were identified, selected and critically reviewed. Keyword: Prostate cancer, Artificial intelligence, prognostics, surgical castration, Treatment response Furthermore, patient’s data features, predictive variables and the techniques of the modellings, performance of the model were extracted and synthesised. The outcome revealed that artificial intelligence prognostic techniques, predictive performances, its clinical applications were able to identify the bases of the disease within the Nigerian contexts. Attention was given to the accuracy of the model, validation as well as the factors which influence the outcome of the treatment. Therefore, artificial intelligence models have the potentials to improve the prediction of the treatment outcome, support the personalised management of metastatic prostate cancer after surgical castration. Although, it is important to established that models validated locally and quality dataset in Nigeria are significant for reliable clinical predictions.