This study presents the design and impleme- -ntation of a Computer-Aided Diagnosis (CAD) system for the automated detection oftumors, fractures, and osteoporosis using machine learning techniques. The study adopted the Convolutional Neural Networks (CNN) for classifying, developing and evaluating medical imaging datasets across the three target pathologies.From the implementat- -ion of the study, the performance of the proposed system reported that the CNN model demonstrated exceptional ability to extract meaningful features from raw medical images, achieving steadily increasing classification accuracy from 72.4% to 93.7% and showing high precision, recall, and F1 scores that reflected strong diagnostic consistency. Precision and recall metrics remained consistently high across all categories, with F1 scores exceeding 0.90 in the test cases from the test dataset. Visualization tools such as confusion matrices and ROC curves confirmed the model’s robustness and generalizability. The CAD system was deployed via a userfriendly web interface, facilitating real-time diagnostic support and enhancing integration into clinical settings. This work contributes an accurate, scalable, and efficient AI-powered diagnostic solution, underscoring the transformative potential of machine learning in medical imaging and intelligent healthcare delivery