Automated classification of a new grading system for diabetic maculopathy based on optical coherence tomography by deep learninng

Abstract

Abstract Purpose: To develop a Vision Transformer model to detect different stages of diabetic maculopathy (DM) based on optical coherence tomography (OCT) images. Methods: A total of 3319 OCT images were extracted from the department of ophthalmology renmin hospital of wuhan university and randomly split the dataset into training and validation sets in a 7:3 ratio. All macular cross-sectional scan OCT images were collected retrospectively from the eyes of DM patients from 2016 to 2022. One of the OCT stages of DM, including early diabetic macular edema (DME), advanced DME, severe DME, and atrophic maculopathy, was labeled on the collected images respectively. A deep learning (DL) model based on Vision Transformer was trained to detect four OCT grading of DM. Results: The model proposed in our paper can provide a detection results effectively. We achieved a mean accuracy of 82.00%, a mean F1 score of 83.11%, a mean AUC of 0.96. The AUC for the detection of four OCT grading (i.e., early DME, advanced DME, severe DME, and atrophic maculopathy) was 0.96, 0.95, 0.87 and 0.98, respectively, with a precision of 88.46%, 80.31%, 89.42% and 87.74%, respectively, a sensitivity of 87.03%, 88.18%, 63.39% and 89.42%, respectively, a specificity of 93.02%, 90.72%, 98.40%, 96.66%, respectively and a F1 score of 87.74%, 84.06%, 88.18% and 88.57%, respectively. Conclusion: Our DL model based on Vision Transformer demonstrated a relatively high accuracy in the detection of OCT grading of DM,, which can help with patients in early screening to obtain a good visual prognosis. These results emphasized the potential of artificial intelligence in assisting clinicians in developing therapeutic strategies with DM in the future .