Dataset Information

"Keep it simple, scholar": an experimental analysis of few-parameter segmentation networks for retinal vessels in fundus imaging.

ABSTRACT:

Purpose

With the recent development of deep learning technologies, various neural networks have been proposed for fundus retinal vessel segmentation. Among them, the U-Net is regarded as one of the most successful architectures. In this work, we start with simplification of the U-Net, and explore the performance of few-parameter networks on this task.

Methods

We firstly modify the model with popular functional blocks and additional resolution levels, then we switch to exploring the limits for compression of the network architecture. Experiments are designed to simplify the network structure, decrease the number of trainable parameters, and reduce the amount of training data. Performance evaluation is carried out on four public databases, namely DRIVE, STARE, HRF and CHASE_DB1. In addition, the generalization ability of the few-parameter networks are compared against the state-of-the-art segmentation network.

Results

We demonstrate that the additive variants do not significantly improve the segmentation performance. The performance of the models are not severely harmed unless they are harshly degenerated: one level, or one filter in the input convolutional layer, or trained with one image. We also demonstrate that few-parameter networks have strong generalization ability.

Conclusion

It is counter-intuitive that the U-Net produces reasonably good segmentation predictions until reaching the mentioned limits. Our work has two main contributions. On the one hand, the importance of different elements of the U-Net is evaluated, and the minimal U-Net which is capable of the task is presented. On the other hand, our work demonstrates that retinal vessel segmentation can be tackled by surprisingly simple configurations of U-Net reaching almost state-of-the-art performance. We also show that the simple configurations have better generalization ability than state-of-the-art models with high model complexity. These observations seem to be in contradiction to the current trend of continued increase in model complexity and capacity for the task under consideration.

SUBMITTER: Fu W

PROVIDER: S-EPMC8166700 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BACKGROUND AND OBJECTIVE:Accurate retinal vessel segmentation is often considered to be a reliable biomarker of diagnosis and screening of various diseases, including cardiovascular diseases, diabetic, and ophthalmologic diseases. Recently, deep learning (DL) algorithms have demonstrated high performance in segmenting retinal images that may enable fast and lifesaving diagnoses. To our knowledge, there is no systematic review of the current work in this research area. Therefore, we performed a systematic review with a meta-analysis of relevant studies to quantify the performance of the DL algorithms in retinal vessel segmentation. METHODS:A systematic search on EMBASE, PubMed, Google Scholar, Scopus, and Web of Science was conducted for studies that were published between 1 January 2000 and 15 January 2020. We followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) procedure. The DL-based study design was mandatory for a study's inclusion. Two authors independently screened all titles and abstracts against predefined inclusion and exclusion criteria. We used the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool for assessing the risk of bias and applicability. RESULTS:Thirty-one studies were included in the systematic review; however, only 23 studies met the inclusion criteria for the meta-analysis. DL showed high performance for four publicly available databases, achieving an average area under the ROC of 0.96, 0.97, 0.96, and 0.94 on the DRIVE, STARE, CHASE_DB1, and HRF databases, respectively. The pooled sensitivity for the DRIVE, STARE, CHASE_DB1, and HRF databases was 0.77, 0.79, 0.78, and 0.81, respectively. Moreover, the pooled specificity of the DRIVE, STARE, CHASE_DB1, and HRF databases was 0.97, 0.97, 0.97, and 0.92, respectively. CONCLUSION:The findings of our study showed the DL algorithms had high sensitivity and specificity for segmenting the retinal vessels from digital fundus images. The future role of DL algorithms in retinal vessel segmentation is promising, especially for those countries with limited access to healthcare. More compressive studies and global efforts are mandatory for evaluating the cost-effectiveness of DL-based tools for retinal disease screening worldwide.

Project description:PurposeThe present study aimed to evaluate the characteristics of fundus autofluorescence in Japanese patients with retinal angiomatous proliferation (RAP).MethodsWe retrospectively reviewed 100 eyes from 76 patients (male, n = 45; female, n = 31; age range, 50-94 years; mean ± standard deviation, 81.4 ± 6.4 years) with treatment-naïve RAP, which was diagnosed based on the identification of retinal-retinal anastomosis on early-phase fluorescein angiography or indocyanine green angiography (ICGA) and the identification of a hot spot on late-phase ICGA. RAP was classified into the following three stages: stage 1, proliferation of intraretinal capillaries originating from the deep retinal complex (intraretinal neovascularization); stage 2, growth of the retinal vessels into the subretinal space (subretinal neovascularization); and stage 3, clinically or angiographically observed choroidal neovascularization. In all cases, short-wavelength and near-infrared autofluorescence (SW-AF, NIR-AF) was evaluated using a confocal scanning laser ophthalmoscope.ResultsThe conditions of the 100 eyes were as follows: stage 1 RAP, n = 6 (6%); stage 2 RAP without retinal pigment epithelial detachment (PED), n = 40 (40%); stage 2 RAP with PED, n = 44 (44%); and stage 3 RAP, 10 (10%). On NIR-AF imaging, the number of abnormalities that were observed to correspond to the RAP lesions on ICGA (87 eyes, 87%) was significantly greater in comparison to SW-AF imaging (27 eyes, 27%). The mean follow-up period in all 76 patients was 39.2 months. In the 76 patients with unilateral disease, 21 (21%) eyes developed RAP in the fellow eye during the follow-up period. Among 18 eyes that were examined by both SW-AF and NIR-AF imaging before the onset of RAP lesions, NIR-AF imaging showed hypoautofluorescence in 15 (83%) eyes before the onset of RAP lesions.ConclusionsSW-AF and NIR-AF abnormalities may be related to the dysfunction of the photoreceptor/retinal pigment epithelium complex. Hypoautofluorescence on NIR-AF imaging may accurately indicate the presence or onset of RAP lesions.