Master’s Thesis Defense on an Intelligent System for Skin Tumor Diagnosis Using Artificial Intelligence at the College of Medicine – Al-Nahrain University

“Skin Tumor Diagnosis Using Deep Learning and Handcrafted Features”

The process began with pre-processing of skin images to improve their quality, followed by feature extraction using deep neural networks along with handcrafted characteristics that precisely describe tissue properties. These data were then input into four classification models, the most prominent being the Support Vector Machine (SVM) model, which achieved the highest accuracy of 97.1%. The evaluation was conducted by splitting the data into 80% for training and 20% for testing, reflecting the reliability of the results and the potential of the proposed system to support medical decision-making in the early detection of skin tumors.