Skip to content

MedVision is an AI-powered web platform for medical researchers, combining tabular data and medical image analysis in one intuitive interface. It automates preprocessing, model training (including 3D U-Net), and evaluation with experiment tracking—designed to accelerate research while maintaining reproducibility.

Notifications You must be signed in to change notification settings

NadaJoobeur/Medical-Research-Automation-Platform

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 

Repository files navigation

🧠 MedVision - Medical Research Automation Platform

Open in GitHub Demo Video Download Project

Web-based platform for end-to-end medical data processing
Supports tabular data & medical images with experiment tracking


🌟 Features

Category Capabilities
Data Import CSV/Excel/DICOM
Visualization 3D slicers, histograms, layer fusion
Preprocessing Auto-normalization, skull-stripping, patch extraction
Model Training U-Net, Inception hybrids, AutoML configs
Evaluation Dice score, ROC curves, confusion matrices
Collaboration Exportable reports, version comparisons

🎥 Interactive Demo

MedVision Demo (Click to see live preprocessing + tabular data treatment demo)



📊 Tech Stack

Component Technology Stack
Frontend React + Redux, Three.js (3D), Plotly.js
Backend Flask, Celery (async tasks), OpenCV/PyDICOM
AI Models TensorFlow/Keras (3D U-Net), Scikit-learn (tabular)
Database MySQL (relational), MinIO (DICOM storage)

🚀 Quick Deployment

1-Click Setup (Docker)

docker-compose -f docker-compose.prod.yml up --build

About

MedVision is an AI-powered web platform for medical researchers, combining tabular data and medical image analysis in one intuitive interface. It automates preprocessing, model training (including 3D U-Net), and evaluation with experiment tracking—designed to accelerate research while maintaining reproducibility.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published