Naveenbio21/RADEXPLAINAI

An offline-first, explainable AI diagnostic platform for radiology, featuring Grad-CAM visual heatmaps and automated clinical reporting. Built for rural and low-resource healthcare environments. #MedicalAI #ExplainableAI #Radiology #Python #HealthcareTech # RadExplainAI: Explainable Diagnostic Assistant for Radiology RadExplainAI is an offline-first, privacy-focused diagnostic platform designed to assist clinicians in rural and low-resource healthcare environments. By leveraging Convolutional Neural Networks (CNNs) and Explainable AI (XAI) frameworks, it provides automated analysis of radiological images accompanied by visual transparency. ## 🩺 Key Features * **Explainable AI (XAI):** Implements Grad-CAM (Gradient-weighted Class Activation Mapping) to generate localized visual heatmaps, allowing clinicians to audit and verify the model's focus areas. * **Offline Autonomy:** Complete local inference pipeline requiring zero cloud connectivity, ensuring 100% data privacy and operational reliability in remote regions. * **Structured Reporting:** Automated generation of multilingual clinical findings to streamline administrative workflows and reporting. * **Forensic Metadata Analysis:** Integrated capabilities for administrative validation and insurance claims adjudication. * **High Portability:** Light footprint optimized for standard desktop hardware and portable, USB-based field deployment. ## 🛠 Tech Stack * **Core Language:** Python * **Deep Learning & Inference:** TensorFlow / PyTorch, `llama-cpp-python` * **Computer Vision & XAI:** OpenCV, Grad-CAM frameworks * **Data Layer:** SQLite * **Export Utilities:** ReportLab (Structured PDF generation) ## ⚠️ Medical Disclaimer **This software is intended for research, educational, and assistive purposes only.** RadExplainAI does not provide definitive medical diagnoses and should never replace the professional judgment, evaluation, or diagnostic verification of a qualified healthcare professional. Developed by Naveenkumar C. | Final Year Biomedical Engineering Student Optimized for the StartupTN & TANSEED Ecosystem ## ⚙️ Setup & Installation 1. **Clone the Repository:** ```bash git clone github.com cd RadExplainAI