Building AR Prenatal Training with Real Data

During my internship at the Indian Institute of Space Science and Technology, I had the opportunity to work on an exciting project that combined my passion for technology with healthcare innovation.

The Challenge

Medical training often lacks hands-on experience with real patient data. Traditional training methods using static images or basic simulations don't provide the immersive experience needed for effective learning.

Our Solution

We developed an AR-based prenatal training tool that:

• Digitalized real ultrasound images and videos for authentic training scenarios
• Created a VR intrauterine environment for immersive learning
• Built a web interface using Cornerstone.js for DICOM image visualization
• Integrated AI capabilities using Gemini API and Vercel AI SDK

Technology Stack

const techStack = {
  frontend: "React",
  3d_modeling: "Unity & Blender", 
  ai_integration: "Gemini API",
  medical_imaging: "Cornerstone.js",
  data_processing: "PyTorch & Python"
};

Key Learnings

1. Medical data handling requires extreme precision and security
2. AR/VR development demands optimization for real-time performance
3. Cross-disciplinary collaboration is essential for healthcare tech

This project was later presented at IEEE GCON Guwahati conference, highlighting the potential of AR in medical education.

Impact

The tool has the potential to revolutionize prenatal training by providing:

• Realistic training scenarios
• Safe learning environment
• Scalable educational solution
• Cost-effective training methodology

This project showcased the intersection of cutting-edge technology and healthcare, demonstrating how AR can enhance medical education.