A Novel Framework for Colon Cancer Detection and Risk Analysis Using Artificial intelligence

Abstract

Early detection of colon cancer plays a crucial role in effective treatment and patient survival. This research focuses on the integration of deep learning techniques with both CT scan images and numerical clinical data for accurate colon cancer detection. The CT scan images were first preprocessed using a mean filter to reduce noise and enhance image quality, ensuring that the affected regions are more distinguishable. Following preprocessing, Convolutional Neural Networks (CNN) were applied for precise segmentation of cancerous regions, while Artificial Neural Networks (ANN) analyzed the numerical patient data to identify risk factors associated with colon cancer. The combination of image-based segmentation and clinical data analysis demonstrates a robust framework for early detection and provides a basis for improved diagnostic accuracy. The results indicate that the use of preprocessing techniques such as the mean filter significantly enhances the performance of deep learning models in detecting colon cancer.