AI-POWERED PROSTATE CANCER DETECTION BY USING CONVOLUTIONAL NEURAL NETWORKS (CNNs)

Electronics and Communication Engineering

This project presents an AI-driven approach for prostate cancer detection and staging using advanced image processing and a Convolutional Neural Network (CNN). Medical images undergo preprocessing steps including resizing, noise reduction, and contrast enhancement to improve uniformity and highlight key features. The CNN then classifies the images as either normal or cancerous, outperforming traditional models like Support Vector Machines (SVM) in both accuracy ([e.g., 95.2%] vs. [e.g., 86.7%]) and inference speed. Unlike SVMs, CNNs automatically extract spatial features, making them more effective for analyzing complex medical images. For images detected as cancerous, a K-means clustering-based segmentation technique isolates tumor regions to assist in cancer staging. This staging step is vital for guiding appropriate treatment plans. The integration of preprocessing, CNN classification, and clustering-based analysis ensures a comprehensive diagnostic system. It reduces manual intervention, speeds up the diagnostic process, and enhances reliability. Overall, the proposed system supports early detection and personalized healthcare in prostate cancer management, marking a significant step forward in AI-assisted diagnostics.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.