Research and Analysis of Research on Automated Segmentation of Microaneurysms in Early Diagnosis of Diabetic Retinopathy

Zhiyong Liu; Chenghe Tang; Jieming Yang

doi:10.54097/7g315w64

Authors

Zhiyong Liu
Chenghe Tang
Jieming Yang

DOI:

https://doi.org/10.54097/7g315w64

Keywords:

Diabetic retinopathy, Microaneurysm, Morphological prior, U-Net.

Abstract

Diabetic retinopathy (DR) is one of the main complications of diabetes, which requires early diagnosis to delay the progression of the disease. As the earliest visible pathological marker of DR, microaneurysm (MA) is of great clinical significance for its precise automatic segmentation. This article systematically reviews the current research progress of MA segmentation technology, focusing on the key challenges in clinical application, such as the small size, low contrast and morphological heterogeneity of MA. On this basis, Ozan Oktay proposed an AG U-Net model that combines an attention mechanism with morphological prior knowledge. By introducing the attention gate control module and dynamic sample weighting strategy, the model significantly improves the accuracy and robustness of segmentation. Experimental results show that the model achieves an excellent dice coefficient of 0.78 on the public data set, which is better than the existing mainstream method while maintaining a high reasoning speed, indicating the promising clinical transformation potential.

References

[1] International Diabetes Federation. IDF Diabetes Atlas, 10th edition. Brussels, Belgium: International Diabetes Federation, 2021.

[2] World Health Organization. World Report on Vision. Geneva: World Health Organization, 2019.

[3] Ranneberger O, Fischer P, & Brox T. U-Net: Convolutional Networks for Biomedical Image Segmentation. In International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI) (pp. 234 – 241). 2015.

[4] Oktay O, Schlemper J, Folgoc L, et al. Attention U-Net: Learning Where to Look for the Pancreas. In Proceedings of the 1st Conference on Medical Imaging with Deep Learning (MIDL). 2018.

[5] Chang J, Wang L, & He J. Study on the Pathogenesis of Diabetic Retinopathy. Advances in Clinical Medicine, 10 (3), 1 - 5. 2020.

[6] Porwal P, Pachade S, Kamble R, et al. Indian Diabetic Retinopathy Image Dataset (IDRiD): A Dataset for Diabetic Retinopathy Segmentation and Grading. IEEE Dataport. 2018.

[7] Decencière E, Zhang X, Cazuguel G, et al. Feedback on a Publicly Distributed Database: The Messidor Database. Image Analysis & Stereology, 33 (3), 231 – 234. 2014.

[8] Adal K M, Etienne C, Ali S, et al Automated Detection of Microaneurysms Using Scale-Adapted Blob Analysis and Semi-Supervised Learning. Computer Methods and Programs in Biomedicine, 114 (1), 1 - 10. 2014.

[9] Li T, Gao Y, Wang K, et al. Diagnostic Assessment of Deep Learning for Diabetic Retinopathy Detection. IEEE Transactions on Medical Imaging, 38 (9), 1 - 1. 2019.

[10] Niemeijer M, Ginneken B V, Cree M J, et al. Retinopathy Online Challenge: Automatic Detection of Microaneurysms in Digital Color Fundus Photographs. IEEE Transactions on Medical Imaging, 29 (1), 185 – 195. 2010.

[11] Tan J H, Acharya U R, Bhandary S V, et al. Segmentation of Optic Disc, Fovea and Retinal Vasculature Using a Single Convolutional Neural Network. Journal of Computational Science, 26, 70 – 79. 2018.

[12] Li X, Hu X, Yu L, et al. MA-Net: A Multi-Scale Attention Network for Liver and Tumor Segmentation. IEEE Access, 8, 179656 – 179665. 2020.

[13] Chen C, Qin C, Qiu H, et al. A Shape-Aware Loss for Medical Image Segmentation. IEEE Transactions on Medical Imaging, 41 (5), 1245 – 1258. 2022.