Medical Images Classification Using Deep learning with Xception Model

Abstract - In the clinical classification of medical images, the Xception model has been seen to be an effective deep learning model for interpreting intricate imaging data. In this research, Xception is utilized to classify chest CT scans into four different classes, including adenocarcinoma, large cell carcinoma, squamous cell carcinoma, and normal lung tissue. With depthwise separable convolutions, the Xception model is efficient at identifying intricate features with less computational overhead. Thorough training, validating, and testing are achieved using well-balanced multiple classes of chest CT datasets to attain comprehensive model evaluation. Classification accuracy across classes with regards to computational efficiency is prioritized in the results. It is seen from the results that Xception is very efficient in differentiating cancer subtypes from normal conditions, thus improving diagnostic consistency. With its feature extraction capabilities, the model has been seen to contribute significantly to better accuracy in medical image classification, promising to have real-world implications for clinical practice and improved patient outcomes.[2] Key Words: Medical image classification, Xception model, Chest CT scan analysis.