Automated Disease Detection from X-ray and CT Scan Images using Deep Learning

This study presents an automated disease detection framework designed to perform binary classification on medical images through a tailored convolution-based neural architecture. Separate models are constructed on X-ray and CT scan datasets, with standardized image preprocessing and regularization techniques applied to enhance data quality and consistency for effective feature extraction. The network is trained using the Adam optimizer combined with categorical cross-entropy loss to effectively differentiate between normal and pathological cases. Model performance is evaluated based on training and validation accuracy, along with corresponding loss values recorded throughout the learning process. Experimental outcomes indicate that the developed CNN delivers reliable classification results, highlighting the practical effectiveness of deep learning techniques for early and automated medical screening.