Recent advancements in artificial intelligence (AI) have led to significant improvements in medical image analysis, especially for early detection and classification of various pathologies. This chapter presents a comprehensive approach to developing a system for automated brain tumor classification using MRI images. Our system, named ISIDA (Intelligent System for Identifying and Diagnosing Anomalies), leverages a convolutional neural network (CNN) for image-based detection and classification, as well as a large language model (LLM) to generate patient-specific diagnostic reports and treatment recommendations. Experimental results on two datasets demonstrate high classification accuracy, even in the presence of noisy images. Additionally, the automated report generation component translates raw classification outputs into meaningful, human-readable text, serving as a useful decision-support tool for medical professionals. This unified approach significantly reduces the workload of healthcare providers, improves diagnostic speed, and offers a scalable solution for medical institutions seeking to integrate AI into their diagnostic workflows.