Abstract:BACKGROUND AND OBJECTIVE: The prevalence of myopia and high myopia is increasing globally, underscoring the growing importance of diagnosing high myopia-related pathologies. While existing image segmentation models, such as U-Net, UNet++, ResU-Net, and TransUNet, have achieved significant success in medical image segmentation, they still face challenges when dealing with ultra-widefield (UWF) fundus images. This study introduces a novel automatic segmentation algorithm for the optic disc and peripapillary atrophy (PPA) based on UWF fundus images, aimed at assisting ophthalmologists in more accurately diagnosing high myopia-related diseases.METHODS: In this study, we developed a segmentation model leveraging a Transformer-based network structure, complemented by atrous convolution and selective boundary aggregation modules, to elevate the accuracy of segmenting the optic disc and PPA in UWF photography. The atrous convolution module adeptly manages multi-scale features, catering to the variances in target sizes and expanding the deep network's receptive field. Concurrently, the incorporation of the selective boundary aggregation module in the skip connections of the model significantly improves the differentiation of boundary information between segmentation targets. Moreover, the comparison of our proposed algorithm with classical segmentation models like U-Net, UNet++, ResU-Net, and TransUNet highlights its considerable advantages in processing UWF photographs.RESULTS: The experimental results show that, compared to the other four models, our algorithm demonstrates substantial improvements in segmenting the optic disc and PPA in UWF photographs. In PPA segmentation, our algorithm improves by 0.8% in Dice, 1.8% in sensitivity, and 1.3% in intersection over union (IOU). In optic disc segmentation, our algorithm improves by 0.3% in Dice, 0.6% in precision, and 0.4% in IOU.CONCLUSION: Our proposed method improves the segmentation accuracy of PPA and optic disks based on UWF photographs, which is valuable for diagnosing high myopia-related diseases in ophthalmology clinics.

Choroidal Atrophy Segmentation Based On Deep Network With Deep-Supervision And Edt-Auxiliary-Loss

Multi-Scale and Multi-Branch Convolutional Neural Network for Retinal Image Segmentation

Choroid Segmentation in OCT Images Based on Improved U-net

A Curriculum Learning-Based Fully Automated System for Quantification of the Choroidal Structure in Highly Myopic Patients

Loss-balanced Parallel Decoding Network for Retinal Fluid Segmentation in OCT.

A new segmentation algorithm for peripapillary atrophy and optic disk from ultra-widefield Photographs1

PGKD-Net: Prior-guided and Knowledge Diffusive Network for Choroid Segmentation

An Efficient Hierarchical Optic Disc and Cup Segmentation Network Combined with Multi-task Learning and Adversarial Learning

Peripapillary Atrophy Segmentation in Fundus Images Via Multi-Task Learning

Application of Artificial Intelligence and Deep Learning for Choroid Segmentation in Myopia

Choroidal layer segmentation in OCT images by a boundary enhancement network

Automatic Segmentation and Visualization of Choroid in OCT with Knowledge Infused Deep Learning

Joint Optic Disc and Cup Segmentation Based on Multi-Module U-shaped Network

Automated segmentation of the optic disc from fundus images using an asymmetric deep learning network

Multi-label deep learning for comprehensive optic nerve head segmentation through data of fundus images

Synergistically segmenting choroidal layer and vessel using deep learning for choroid structure analysis

MIU-Net: Advanced Multi-Scale Feature Extraction and Imbalance Mitigation for Optic Disc Segmentation

Automated segmentation of macular edema for the diagnosis of ocular disease using deep learning method

Segmentation of RBCC Disruption and Myopic Stretch Line in Retinal OCT Images Using an Improved U-shape Network

A New Convolutional Neural Network Model for Peripapillary Atrophy Area Segmentation from Retinal Fundus Images

Automated Segmentation of the Optic Disc Using the Deep Learning.