Convex Shapes (CS) are common priors for optic disc and cup segmentation in eye fundus images. It is important to design proper techniques to represent convex shapes. So far, it is still a problem to guarantee that the output objects from a Deep Neural Convolution Networks (DCNN) are convex shapes. In this work, we propose a technique which can be easily integrated into the commonly used DCNNs for image segmentation and guarantee that outputs are convex shapes. This method is flexible and it can handle multiple objects and allow some of the objects to be convex. Our method is based on the dual representation of the sigmoid activation function in DCNNs. In the dual space, the convex shape prior can be guaranteed by a simple quadratic constraint on a binary representation of the shapes. Moreover, our method can also integrate spatial regularization and some other shape prior using a soft thresholding dynamics (STD) method. The regularization can make the boundary curves of the segmentation objects to be simultaneously smooth and convex. We design a very stable active set projection algorithm to numerically solve our model. This algorithm can form a new plug-and-play DCNN layer called CS-STD whose outputs must be a nearly binary segmentation of convex objects. In the CS-STD block, the convexity information can be propagated to guide the DCNN in both forward and backward propagation during training and prediction process. As an application example, we apply the convexity prior layer to the retinal fundus images segmentation by taking the popular DeepLabV3+ as a backbone network. Experimental results on several public datasets show that our method is efficient and outperforms the classical DCNN segmentation methods.
Detecting diabetic retinopathy severity through fundus images using an ensemble of classifiers
Technical University of Cluj-Napoca（克卢日-纳波卡技术大学）；“Iuliu Hatieganu” University of Medicine and Pharmacy（“尤利乌·哈采加努”医药大学）；Emergency County Hospital（急诊县医院）
Fully Convolutional Neural Network for Semantic Segmentation of Anatomical Structure and Pathologies in Colour Fundus Images Associated with Diabetic Retinopathy
Indian Institute of Technology Kharagpur（印度理工学院卡拉格普尔分校）
DoFE: Domain-oriented Feature Embedding for Generalizable Fundus Image Segmentation on Unseen Datasets
The Chinese University of Hong Kong（香港中文大学）；Stanford University（斯坦福大学）；Shenzhen University（深圳大学）；Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences（中国科学院深圳先进技术研究院）
Cost-Sensitive Regularization for Diabetic Retinopathy Grading from Eye Fundus Images
University of Bournemouth（伯恩茅斯大学）；École de Technolgie Superieure de Montréal（蒙特利尔高等技术学院）；Diagnos Inc.（Diagnos公司）；CRCHUM University of Montreal Hospital Centre（蒙特利尔大学医院研究中心）
Simultaneous segmentation and classification of the retinal arteries and veins from color fundus images
Centro de Investigación CITIC（CITIC研究中心）；Universidade da Coruña（拉科鲁尼亚大学）；Instituto de Investigación Biomédica de A Coruña (INIBIC)（拉科鲁尼亚生物医学研究所（INIBIC））
Learned Pre-Processing for Automatic Diabetic Retinopathy Detection on Eye Fundus Images
Carnegie Mellon University（卡内基梅隆大学）；INESC TEC, Porto（波尔图INESC TEC研究所）；École de Tecnologie Superieure, Montreal（蒙特利尔高等技术学院）；Faculty of Engineering, University of Porto（波尔图大学工程学院）
Superimposition of eye fundus images for longitudinal analysis from large public health databases
International Prevention Research Institute（国际预防研究所）；Swansea University（斯旺西大学）；Diabetic Eye Screening Wales（威尔士糖尿病眼部筛查机构）；University of Strathclyde（思克莱德大学）
SSVT: Self-Supervised Vision Transformer For Eye Disease Diagnosis Based On Fundus Images
Beihang University（北京航空航天大学）；Beijing Tongren Hospital, Capital Medical University（首都医科大学附属北京同仁医院）；The University of Melbourne（墨尔本大学）；University of Cambridge（剑桥大学）；Imperial College London（帝国理工学院）
ODFormer: Semantic Fundus Image Segmentation Using Transformer for Optic Nerve Head Detection
College of Electronics & Information Engineering, Tongji University（同济大学电子与信息工程学院）；Shanghai Institute of Intelligent Science and Technology, Tongji University（同济大学上海智能科学与技术研究院）；School of Life Science, Shanghai University（上海大学生命科学学院）；Tongji Hospital, School of Medicine, Tongji University（同济大学医学院附属同济医院）
Automated Artifact Detection in Ultra-widefield Fundus Photography of Patients with Sickle Cell Disease
Johns Hopkins School of Medicine（约翰斯·霍普金斯医学院）；Wilmer Eye Institute（威尔眼科研究所）；Morehouse School of Medicine（莫尔豪斯医学院）；Drexel University College of Medicine（德雷塞尔大学医学院）；Malone Center for Engineering in Healthcare（马龙医疗保健工程中心）
Fundus2Globe: Generative AI-Driven 3D Digital Twins for Personalized Myopia Management
School of Optometry, The Hong Kong Polytechnic University（香港理工大学验光学院）；Swiss Federal Institute of Technology Lausanne (EPFL)（瑞士洛桑联邦理工学院）；Centre for Eye and Vision Research (CEVR)（眼与视觉研究中心）
DeepSeeNet: A deep learning model for automated classification of patient-based age-related macular degeneration severity from color fundus photographs
National Center for Biotechnology Information (NCBI)（美国国家生物技术信息中心）；National Library of Medicine (NLM)（美国国家医学图书馆）；National Institutes of Health (NIH)（美国国立卫生研究院）；National Eye Institute (NEI)（美国国家眼科研究所）
Identifying the key components in ResNet-50 for diabetic retinopathy grading from fundus images: a systematic investigation
Southern University of Science and Technology（南方科技大学）；The University of British Columbia（不列颠哥伦比亚大学）；The University of Hong Kong（香港大学）；The University of Queensland（昆士兰大学）
A deep learning approach for automated detection of geographic atrophy from color fundus photographs
National Eye Institute (NEI), National Institutes of Health (NIH)（美国国立卫生研究院国立眼科研究所）；National Center for Biotechnology Information (NCBI), National Library of Medicine (NLM), National Institutes of Health (NIH)（美国国立卫生研究院国家医学图书馆国家生物技术信息中心）
Versatile Cataract Fundus Image Restoration Model Utilizing Unpaired Cataract and High-quality Images
Shenzhen-Platform and Carrier (International Science and Technology Information Center) & Shenzhen Bay Lab（深圳平台与载体（国际科技信息中心）及深圳湾实验室）；Capital Medical University（首都医科大学）；Beijing Tongren Hospital（北京同仁医院）
MultiEYE: Dataset and Benchmark for OCT-Enhanced Retinal Disease Recognition from Fundus Images
Hong Kong University of Science and Technology（香港科技大学）；Guangdong Weiren Meditech Co., Ltd.（广东伟仁医疗科技有限公司）；Yunnan United Vision Innovations Technology Co., Ltd.（云南联合视力创新科技有限公司）；Guangdong Hospital of Integrated Traditional Chinese and Western Medicine（广东省中西医结合医院）；Second People’s Hospital of Foshan（佛山市第二人民医院）
Deep Learning and Glaucoma Specialists: The Relative Importance of Optic Disc Features to Predict Glaucoma Referral in Fundus Photos
Google Health（谷歌健康）；Google LLC（谷歌有限责任公司）；Virginia Ophthalmology Associates（弗吉尼亚眼科协会）；Eastern Virginia Medical School（东弗吉尼亚医学院）；Duke University（杜克大学）；Emory University School of Medicine（埃默里大学医学院）；Atlanta Veterans Affairs Medical Center（亚特兰大退伍军人事务医疗中心）；Dr. Shroff’s Charity Eye Hospital（施罗夫博士慈善眼科医院）
Deep Learning for Automated Quality Assessment of Color Fundus Images in Diabetic Retinopathy Screening
Australian e-Health Research Centre, CSIRO（澳大利亚电子健康研究中心，联邦科学与工业研究组织）；ACRV, The Australian National University（澳大利亚机器人视觉中心，澳大利亚国立大学）；University of California, Berkeley（加州大学伯克利分校）
INESC TEC（系统与计算机工程研究所 - 科技与科学）；Stephen A Wynn Institute for Vision Research（Stephen A Wynn视觉研究所）；University of Iowa Hospital and Clinics（爱荷华大学医院和诊所）；IDx LLC（IDx有限责任公司）；University of Porto（波尔图大学）
Explainable Convolutional Neural Networks for Retinal Fundus Classification and Cutting-Edge Segmentation Models for Retinal Blood Vessels from Fundus Images
Ahsanullah University of Science and Technology（阿萨努拉科技大学）
