Artificial Intelligence (AI) technologies have improved their ability to recognise complicated patterns in image processing in recent years. In healthcare technology, AI is playing an important role in detecting illnesses, monitoring patients, and expanding the medical industry. Medical Image Processing (MIP) is the most difficult and rapidly increasing subject of computer science. Image segmentation, object recognition, feature extraction, and corner edge detection in MRI images are some of the key issues in image processing jobs. Traditional filter-based edge detection techniques gain from theoretical advances, but processing requires more calculations. The primary objective of this research endeavour is to devise a new technique for enhancing the precision of heart disease identification in cardiac MRI images through the utilisation of Convolutional Neural Networks (CNN). The proposed methodology entails an enhanced Canny edge detection technique that leverages Convolutional Neural Networks (CNNs). The proposed methodology involves detecting the pixel values of the corners of MRI images while scanning in the opposite direction, resulting in the omission of adjacent data, and attaining a precision rate of 98.98%. The results point to more reliable identification of edge feature sets by MRI image edge detection.

Hear disease, canny edge detection, tabu search, PSO, and CNN.