NS
Nishant OS
Initializing Nishant AI
LOADING0%
All work
ResearchMedical Computer Vision · Deep Learning · Sep 2025 – Jan 2026

Diabetic Retinopathy Detection

Grading diabetic retinopathy from retinal fundus images at clinical agreement.

90.42%
Accuracy
5-class
0.94
Quadratic Kappa
clinical agreement
0
Backbones
EfficientNet · ConvNeXt · Swin
Grad-CAM
Explainability

A deep-learning pipeline for multi-class grading of diabetic retinopathy from retinal fundus images, benchmarking EfficientNet, ConvNeXt, and Swin-Tiny backbones and adding Grad-CAM explainability to highlight clinically relevant regions. It reaches 90.42% accuracy and a 94.17% Quadratic Weighted Kappa — indicating strong agreement with clinical grading.

The problem

Diabetic retinopathy is a leading cause of preventable blindness, but screening is bottlenecked by the shortage of ophthalmologists — and black-box models won't earn clinical trust.

The solution

Engineered pipelines across three modern backbones (EfficientNet, ConvNeXt, Swin-Tiny) for five-class DR classification, paired with Grad-CAM visualizations so predictions can be inspected against clinically relevant regions of the retina.

System Architecture
Fundus ImagePreprocesscrop · normalizeBackbonesEfficientNet · ConvNeXt · Swin5-Class GradeGrad-CAMexplainability
Challenges solved
  • 1Multi-class imbalance across the five DR severity grades.
  • 2Interpretability strong enough to earn clinical trust.
  • 3Benchmarking modern CNN and transformer backbones fairly.
Highlights
  • 90.42% accuracy with 0.94 Quadratic Weighted Kappa.
  • Three-backbone benchmark: EfficientNet, ConvNeXt, Swin-Tiny.
  • Grad-CAM overlays for clinical interpretability.
Technology
PythonPyTorchTensorFlowEfficientNetConvNeXtSwin-TinyGrad-CAMOpenCV
Next project
DRISHTI
Assistive AI · Edge · IoT