Development of automatic retinal vessel segmentation method in fundus images via convolutional neural networks

Retinal Vessel Segmentation With Skeletal Prior and Contrastive Loss

The morphology of retinal vessels is closely associated with many kinds of ophthalmic diseases. Although huge progress in retinal vessel segmentation has been achieved with the advancement of deep learning, some challenging issues remain. For example, vessels can be disturbed or covered by other components presented in the retina (such as optic disc or lesions). Moreover, some thin vessels are also easily missed by current methods. In addition, existing fundus image datasets are generally tiny, due to the difficulty of vessel labeling. In this work, a new network called SkelCon is proposed to deal with these problems by introducing skeletal prior and contrastive loss. A skeleton fitting module is developed to preserve the morphology of the vessels and improve the completeness and continuity of thin vessels. A contrastive loss is employed to enhance the discrimination between vessels and background. In addition, a new data augmentation method is proposed to enrich the training samples and improve the robustness of the proposed model. Extensive validations were performed on several popular datasets (DRIVE, STARE, CHASE, and HRF), recently developed datasets (UoA-DR, IOSTAR, and RC-SLO), and some challenging clinical images (from RFMiD and JSIEC39 datasets). In addition, some specially designed metrics for vessel segmentation, including connectivity, overlapping area, consistency of vessel length, revised sensitivity, specificity, and accuracy were used for quantitative evaluation. The experimental results show that, the proposed model achieves state-of-the-art performance and significantly outperforms compared methods when extracting thin vessels in the regions of lesions or optic disc. Source code is available at https://www.github.com/tyb311/SkelCon.

Retinal Vessel Segmentation, a Review of Classic and Deep Methods

Retinal illnesses such as diabetic retinopathy (DR) are the main causes of vision loss. In the early recognition of eye diseases, the segmentation of blood vessels in retina images plays an important role. Different symptoms of ocular diseases can be identified by the geometric features of ocular arteries. However, due to the complex construction of the blood vessels and their different thicknesses, segmenting the retina image is a challenging task. There are a number of algorithms that helped the detection of retinal diseases. This paper presents an overview of papers from 2016 to 2022 that discuss machine learning and deep learning methods for automatic vessel segmentation. The methods are divided into two groups: Deep learning-based, and classic methods. Algorithms, classifiers, pre-processing and specific techniques of each group is described, comprehensively. The performances of recent works are compared based on their achieved accuracy in different datasets in inclusive tables. A survey of most popular datasets like DRIVE, STARE, HRF and CHASE_DB1 is also given in this paper. Finally, a list of findings from this review is presented in the conclusion section.

A Coarse-to-Fine Fully Convolutional Neural Network for Fundus Vessel Segmentation

Fundus vessel analysis is a significant tool for evaluating the development of retinal diseases such as diabetic retinopathy and hypertension in clinical practice. Hence, automatic fundus vessel segmentation is essential and valuable for medical diagnosis in ophthalmopathy and will allow identification and extraction of relevant symmetric and asymmetric patterns. Further, due to the uniqueness of fundus vessel, it can be applied in the field of biometric identification. In this paper, we remold fundus vessel segmentation as a task of pixel-wise classification task, and propose a novel coarse-to-fine fully convolutional neural network (CF-FCN) to extract vessels from fundus images. Our CF-FCN is aimed at making full use of the original data information and making up for the coarse output of the neural network by harnessing the space relationship between pixels in fundus images. Accompanying with necessary pre-processing and post-processing operations, the efficacy and efficiency of our CF-FCN is corroborated through our experiments on DRIVE, STARE, HRF and CHASE DB1 datasets. It achieves sensitivity of 0.7941, specificity of 0.9870, accuracy of 0.9634 and Area Under Receiver Operating Characteristic Curve (AUC) of 0.9787 on DRIVE datasets, which surpasses the state-of-the-art approaches.

The Relationship of Retinal Vessel Geometric Characteristics to the Incidence and Progression of Diabetic Retinopathy

PURPOSE

To examine the relationships of retinal vessel geometric characteristics (RVGCs) to the incidence and progression of diabetic retinopathy (DR).

DESIGN

Observational, prospective cohort study.

PARTICIPANTS

Nine hundred ninety-six persons with type 1 diabetes mellitus (T1DM) and 1370 persons with type 2 diabetes mellitus (T2DM) seen at a baseline examination who were eligible for follow-up examinations at subsequent 5-year intervals. A total of 3846 person-interval data from these follow-up examinations are the basis for the analyses.

METHODS

Diabetic retinopathy and macular edema were assessed by grading of 30° stereoscopic color fundus photographs. Retinal vessel geometric characteristics were assessed using the Singapore I Vessel Assessment program from a digitized copy of 1 of the field 1 fundus photographs obtained at baseline and follow-up.

MAIN OUTCOME MEASURES

The 5-year incidence of any DR, progression of DR, and incidence of proliferative diabetic retinopathy (PDR) and clinically significant macular edema (CSME) in right eyes.

RESULTS

Incident DR occurred in 45%, progression in 32%, PDR in 10%, and CSME in 5%. While adjusting for glycated hemoglobin, duration of diabetes, and other factors, retinal arteriolar simple tortuosity was associated significantly with the incidence of any DR (odds ratio [OR], 1.17; 95% confidence interval [CI], 1.01-1.35). Retinal venular branching angle was associated significantly with progression of DR (OR, 1.18; 95% CI, 1.03-1.36), retinal venular curvature tortuosity was associated significantly with the incidence of PDR (OR, 1.15; 95% CI, 1.01-1.30), and retinal venular branching angle (OR, 1.41; 95% CI, 1.10-1.82) was associated significantly with the incidence of CSME. There were no significant associations of other RVGCs with any of the DR outcomes in the full multivariate model. Inclusion of all possible RVGCs did not improve the predictive value of the models that already included retinal vessel diameter and baseline DR severity level.

CONCLUSIONS

Retinal vessel geometric characteristics of the retinal venules were associated with progression of DR; however, most of the RVGCs measured from digitized fundus photographs added little to the assessment of risk of incidence and progression of DR when other risk factors were considered in T1DM and T2DM.