Abstract
Prostate cancer is one of the most common cancers that is detected in males and is diagnosed with difficulty in the early stages. This paper uses advances in machine learning to develop an integrated framework combining clinical data and medical imaging for predicting prostate cancer. Using the advanced preprocessing of class balancing and standardization of features with multi-models of ML algorithms applied for better diagnostic accuracy and scalability, the study attempts to improve. The classical algorithms include k-Nearest Neighbors (KNN) and Naive Bayes. Convolutional Neural Networks (CNN) have also been utilized to compare deep learning to determine imaging features from 3D MRI scans. This gives the peak test accuracy to CNNs at 88%, showing strong superiority over traditional methods in the process. It depicts the potential of AI tools in changing the method of diagnostic workflows so that it can be done non-invasively and achieve more efficient clinical decision-making tools.