Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{197319,
        author = {SHANTANU VIKAS CHARPE and DEVENDRA TAYADE and ANUP URKUNDE and PROF. ARTI BHURGHATE and DIVYANSH TIWARI and SAVIDHAN KHANDARE},
        title = {AI MEDICAL IMAGE DIAGNOSIS},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {6498-6507},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=197319},
        abstract = {Artificial Intelligence (AI) has shown great promise in helping doctors diagnose diseases from medical images like MRI scans. However, most AI research focuses only on making accurate predictions, ignoring the practical needs of real hospitals—keeping patient records, generating reports, and working without internet. This paper presents MedAI, a complete system that solves this problem. MedAI uses a powerful but efficient AI model called MobileNetV2 to classify brain MRI images into four categories: Glioma (a type of brain tumor), Meningioma (another brain tumor), Pituitary tumor, or Normal (healthy). What makes MedAI special is that it does much more than just prediction. It stores every diagnosis in a local database linked to patient information, shows doctors an interactive dashboard with useful analytics, and generates professional PDF reports that can be printed or saved. The entire system works offline, requires no cloud services, and is designed for use in clinics with limited resources. The AI model achieves high confidence scores (typically above 90% for correct predictions) by using transfer learning—starting from a model already trained on millions of everyday images and then fine-tuning it on brain MRI data. The system is built with four layers: (1) Data Ingestion (automatically reads images from folders), (2) AI Engine (makes predictions), (3) Data Persistence (saves everything to SQLite database), and (4) Clinical Interface (Streamlit dashboard + PDF reports). This paper provides complete implementation details, architecture diagrams, and performance metrics. MedAI represents a practical, deployable solution that bridges the gap between AI research and real clinical use.},
        keywords = {Medical Image Classification, Brain MRI, Deep Learning, MobileNetV2, Transfer Learning, Clinical Dashboard, SQLite, Streamlit, HealthcareAI},
        month = {April},
        }