Diabetic Retinopathy (DR) has become one of the leading causes of vision impairment in working-aged people and is a severe problem worldwide. However, most of the works ignored the ordinal information of labels. In this project, we propose a novel design MTCSNN, a Multi-task Clinical Siamese Neural Network for Diabetic Retinopathy severity prediction task. The novelty of this project is to utilize the ordinal information among labels and add a new regression task, which can help the model learn more discriminative feature embedding for fine-grained classification tasks. We perform comprehensive experiments over the RetinaMNIST, comparing MTCSNN with other models like ResNet-18, 34, 50. Our results indicate that MTCSNN outperforms the benchmark models in terms of AUC and accuracy on the test dataset.
@article{arxiv.2208.06917,
title = {MTCSNN: Multi-task Clinical Siamese Neural Network for Diabetic Retinopathy Severity Prediction},
author = {Chao Feng and Jui Po Hung and Aishan Li and Jieping Yang and Xinyu Zhang},
journal= {arXiv preprint arXiv:2208.06917},
year = {2022}
}
Comments
This paper is not sufficiently exhaustive and lacks some analysis. Besides, certain methods of this paper are from the first author's other co-first authoring research paper. There exist disputes among authors, thus we decide to withdraw this paper currently