/Retinal-Disease-Classification-using-Optical-Coherence-Topographical-Scans

Retinal Disease Classification using Optical Coherence Topographical Scans

Primary LanguageJupyter NotebookMIT LicenseMIT

Retinal-Disease-Classification-using-Optical-Coherence-Topographical-Scans

This is a multiclass classification of different retinal disease like-

  1. Choroidal neovascularization (CNV)
  2. Diabetic macular edema (DME)
  3. Drusen
  4. Normal Condition.

Using advanced image processing and deep learning algorithms the results of the classification are-

  1. Training Accuracy = 99.91 %

  2. Validation Accuracy = 95.92 %

  3. Test Accuracy = 96.52 %

  4. Sensitivity = 99.13 %

  5. Specificity = 97.95 %

Read the paper at https://ieeexplore.ieee.org/document/9276708

Run in your machine:

  • git clone https://github.com/sudo-rajarshi/Retinal-Disease-Classification-using-Optical-Coherence-Topographical-Scans.git
  • cd Retinal-Disease-Classification-using-Optical-Coherence-Topographical-Scans
  • pip3 install -r requirements.txt
  • Open jupyter notebook and enjoy!