scanmymri.ai converts raw DICOM MRI scan folders into interactive 3D visualizations using the power of Dash and VTK. Users can upload a full MRI series—such as T1-weighted axial slices—and the application automatically parses the DICOM metadata, reconstructs the volume, and renders a high-resolution 3D model directly in the browser. Built on Dash VTK, the viewer allows users to rotate, zoom, and pan the brain model, as well as switch between axial, sagittal, and coronal slice planes. Window/level adjustments, color maps, and opacity controls provide a more refined visualization experience. What sets this tool apart is its integration with GPT-4.1-mini (or GPT-4o), which allows users to run AI-assisted analysis on the MRI data. With a single click, the model can highlight anatomical structures, flag potential abnormalities, and answer natural-language questions about the scan. It can also generate clinical-style summaries or simplified explanations tailored to medical students or patients. Whether you're a radiology trainee exploring neuroanatomy or a researcher working with public datasets like BraTS or TCIA, the platform offers an intuitive and intelligent way to analyze MRI volumes. The backend is modular and extensible, built with Python and Dash, and supports preprocessing steps like resampling and normalization. The roadmap includes features such as multimodal image support (CT, PET), region-specific AI annotation, quantitative volumetric metrics, and even STL export for 3D printing. Future updates aim to make the platform HIPAA-compliant for potential clinical use. By combining advanced visualization with interactive AI, this tool bridges the gap between complex medical imaging and user-friendly, intelligent interpretation.
