Case Study: Diabetic Retinopathy Detection Powered by AI

FPT Software developed the solution using machine learning/deep learning (AI) to identify and diagnose lung cancer, skin diseases, muscle and fat measurement, and more through CT images.

Industry: Healthcare

The Solution

  • Medical image processing
    • Filtering
    • Segmentation
    • Bone suppression
    • 3D/4D image processing
  • Deep learning/machine learning
    • Image segmentation
    • Imaging diagnosis
    • Medical image quality improvement

Features

  • Define problems from customers and propose solutions
  • Implement PoC and evaluate and improve solutions

Business Needs

  • Diabetic retinopathy (DR) is a long-term diabetes associate with eye disease.
  • If DR can be detected within a certain period of time, it can delay or avoid progression to vision impairment.

Scope and Results

  • Provide optimal deep learning model
  • Add position information to that model and do analysis to improve accuracy
  • Propose new method for acquiring and labeling learning data.

Technology

TensorFlow 1.5 (AI), OpenCV 3.4, MaskRCNN, KNN

Diabetic Retinopathy Detection Powered by AI