CNN Model for Classifying Age-Related Macular Degeneration Using OCT
Author Information
Author(s): Yusufoğlu Elif, Fırat Hüseyin, Üzen Hüseyin, Özçelik Salih Taha Alperen, Çiçek İpek Balıkçı, Şengür Abdulkadir, Atila Orhan, Guldemir Numan Halit
Primary Institution: Fırat University
Hypothesis
Can a novel CNN-based deep learning method improve the diagnosis of age-related macular degeneration (AMD) using optical coherence tomography (OCT) images?
Conclusion
The study demonstrates that the proposed CNN model achieved high accuracy in classifying AMD, indicating its potential for improving early detection and intervention.
Supporting Evidence
- The proposed model achieved 97.98% accuracy on a private dataset and 100% accuracy on a public dataset.
- The model outperformed existing deep learning approaches in classifying AMD.
- Key performance metrics included 97.95% precision, 97.77% recall, and 97.86% F1 score on the private dataset.
- The study emphasizes the importance of AI in enhancing diagnostic accuracy for AMD.
Takeaway
This study created a computer program that helps doctors find eye problems in older people by looking at special pictures of their eyes. It works really well!
Methodology
The study used a CNN model that combines modified Inception modules, Depthwise Squeeze-and-Excitation Blocks, and ConvMixer architecture to analyze OCT images.
Limitations
The study is limited by the size and diversity of the datasets used for training and testing.
Participant Demographics
The private dataset included 256 participants with a total of 2316 OCT images from 653 eyes.
Statistical Information
P-Value
p<0.05
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website