Light-weight Retinal Layer Segmentation with Global Reasoning
School of Mechanical Engineering, Shandong University(山东大学机械工程学院);Joint SDU-NTU Centre for Artificial Intelligence Research (C-FAIR)(山东大学-南洋理工大学人工智能联合研究中心);Fondazione Bruno Kessler(布鲁诺·凯塞勒基金会)
Xiang He, Weiye Song, Yiming Wang, Fabio Poiesi, Ji Yi, Manishi Desai, Quanqing Xu, Kongzheng Yang, Yi Wan
备注:
IEEE Transactions on Instrumentation & Measurement
Automatic retinal layer segmentation with medical images, such as optical coherence tomography (OCT) images, serves as an important tool for diagnosing ophthalmic diseases. However, it is challenging to achieve accurate segmentation due to low contrast and blood flow noises presented in the images. In addition, the algorithm should be light-weight to be deployed for practical clinical applications. Therefore, it is desired to design a light-weight network with high performance for retinal layer segmentation. In this paper, we propose LightReSeg for retinal layer segmentation which can be applied to OCT images. Specifically, our approach follows an encoder-decoder structure, where the encoder part employs multi-scale feature extraction and a Transformer block for fully exploiting the semantic information of feature maps at all scales and making the features have better global reasoning capabilities, while the decoder part, we design a multi-scale asymmetric attention (MAA) module for preserving the semantic information at each encoder scale. The experiments show that our approach achieves a better segmentation performance compared to the current state-of-the-art method TransUnet with 105.7M parameters on both our collected dataset and two other public datasets, with only 3.3M parameters.
Automated segmentation of retinal fluid volumes from structural and angiographic optical coherence tomography using deep learning
Casey Eye Institute, Oregon Health & Science University(俄勒冈健康与科学大学凯西眼科研究所);School of Physics and Optoelectronic Engineering, Foshan University(佛山大学物理与光电工程学院);Department of Biomedical Engineering, Oregon Health & Science University(俄勒冈健康与科学大学生物医学工程系)
LVS-Net: A Lightweight Vessels Segmentation Network for Retinal Image Analysis
School of Electronics, Electrical Engineering and Computer Science, Queen’s University Belfast(贝尔法斯特女王大学电子、电气工程与计算机科学学院);Abasyn University(阿巴辛大学);School of Computer Science and Engineering, University of New South Wales(新南威尔士大学计算机科学与工程学院)
ROSE: A Retinal OCT-Angiography Vessel Segmentation Dataset and New Model
Cixi Institute of Biomedical Engineering, Ningbo Institute of Industrial Technology, Chinese Academy of Sciences(中国科学院宁波工业技术研究院慈溪生物医学工程研究所);Inception Institute of Artificial Intelligence(起源人工智能研究院);Keck School of Medicine, University of Southern California(南加州大学凯克医学院);First Affiliated Hospital of Wenzhou Medical University(温州医科大学附属第一医院);Southern University of Science and Technology(南方科技大学);University of Liverpool(利物浦大学)
MPG-Net: Multi-Prediction Guided Network for Segmentation of Retinal Layers in OCT Images
University of Oxford(牛津大学);Big Data Institute, University of Oxford(牛津大学大数据研究所);School of Engineering, Newcastle University(纽卡斯尔大学工程学院);Neura Health Group;Newcastle Eye Centre, Royal Victoria Infirmary(纽卡斯尔皇家维多利亚医院纽卡斯尔眼科中心);Newcastle University(纽卡斯尔大学);Newcastle Upon Tyne, UK(泰恩河畔纽卡斯尔(英国))
Simultaneous Alignment and Surface Regression Using Hybrid 2D-3D Networks for 3D Coherent Layer Segmentation of Retinal OCT Images with Full and Sparse Annotations
Xiamen University(厦门大学);National Institute for Data Science in Health and Medicine, Xiamen University(厦门大学健康医疗大数据国家研究院);Tencent YouTu Lab(腾讯优图实验室);Southern University of Science and Technology(南方科技大学)
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)
Segmentation-free integration of nuclei morphology and spatial transcriptomics for retinal images
Dept. of Information Technology(信息技术系);Dept. of Molecular, Cell and Developmental Biology(分子、细胞与发育生物学系);University of California, Los Angeles(加利福尼亚大学洛杉矶分校)
ReLayNet: Retinal Layer and Fluid Segmentation of Macular Optical Coherence Tomography using Fully Convolutional Network
Technische Universität München(慕尼黑工业大学);Ludwig-Maximilians-University(路德维希-马克西米利安大学);Indian Institute of Technology Kharagpur(印度理工学院克勒格布尔分校);Johns Hopkins University(约翰斯·霍普金斯大学)
Parametric Scaling of Preprocessing assisted U-net Architecture for Improvised Retinal Vessel Segmentation
Siksha ‘O’ Anusandhan (Deemed to be University)(锡克沙奥阿努山丹大学(被认定为大学));ITER, Siksha ‘O’ Anusandhan (Deemed to be University)(锡克沙奥阿努山丹大学(被认定为大学)ITER学院);Center for the Internet of Things, ITER, Siksha ‘O’ Anusandhan (Deemed to be University)(锡克沙奥阿努山丹大学(被认定为大学)ITER学院物联网中心)
SD-RetinaNet: Topologically Constrained Semi-Supervised Retinal Lesion and Layer Segmentation in OCT
Christian Doppler Laboratory for Artificial Intelligence in Retina(基督教多普勒人工智能视网膜实验室);Institute of Artificial Intelligence(人工智能研究所);Center for Medical Data Science(医学数据科学中心);Medical University of Vienna(维也纳医学大学);Computational Imaging Research Lab(计算成像研究实验室);Department of Biomedical Imaging and Image-Guided Therapy(生物医学成像与影像引导治疗部门);Department of Ophthalmology and Optometry(眼科与视光学部门)
Retinal Layer Segmentation in OCT Images With 2.5D Cross-slice Feature Fusion Module for Glaucoma Assessment
Department of Mechanical Engineering, Yonsei University, Seoul, 03722, South Korea(延世大学机械工程系);Institute of Vision Research, Department of Ophthalmology, Severance Hospital, Yonsei University, Seoul, 03722, South Korea(视觉研究所眼科部, 顺化医院, 延世大学, 首尔, 03722, 韩国);Yonsei University College of Medicine, Seoul, 03722, South Korea(延世大学医学院)
Exploiting Epistemic Uncertainty of Anatomy Segmentation for Anomaly Detection in Retinal OCT
Medical University Vienna(维也纳医科大学);University of Vienna(维也纳大学);NVIDIA Corporation(英伟达公司)
通过检测医学图像中的相关生物标志物来辅助诊断和治疗指导。虽然监督式深度学习可以对病理区域进行准确分割,但它受到限制,需要对这些区域进行先验定义,大规模注释以及训练集中的代表性患者队列。相反,异常检测不限于病理的特定定义,并且允许在没有注释的情况下对健康样本进行训练。然后,异常区域可以作为生物标志物发现的候选者。关于正常解剖结构的知识带来了用于检测异常的隐含信息。我们建议使用baidu深度学习来利用这一特性,基于这样的假设,即认知的不确定性将与正常训练集的解剖偏差相关。贝叶斯U形网是在一个定义良好的健康环境中训练的,使用现有方法产生的健康解剖结构的弱标签。在测试时,我们使用Monte Carlo dropout捕获模型的认知不确定性估计。一种新的后处理技术,然后应用到利用这些估计,并将其分层的外观平滑斑点形状的分割的异常。我们实验验证了这种方法在视网膜光学相干断层扫描(OCT)图像,使用弱标签的视网膜层。我们的方法在年龄相关性黄斑变性(AMD)病例的独立异常测试集中获得了0.789的Dice指数。所得到的分割允许非常高的准确性,以分离健康和患病的情况下,晚期湿性AMD,干地图状萎缩(GA),糖尿病性黄斑水肿(DME)和视网膜静脉阻塞(RVO)。最后,我们定性地观察到,我们的方法也可以检测正常扫描中的其他偏差,如切割边缘伪影。
Learning Mutually Local-global U-nets For High-resolution Retinal Lesion Segmentation in Fundus Images
The Chinese University of Hong Kong (Shenzhen)(香港中文大学(深圳));Shenzhen Research Institute of Big Data(深圳大数据研究院);University of Science and Technology of China(中国科学技术大学)
Universal Wavelet Units in 3D Retinal Layer Segmentation
Jacobs School of Engineering, University of California San Diego(加州大学圣地亚哥分校 Jacobs 工程学院);Jacobs Retina Center, Shiley Eye Institute, University of California San Diego(加州大学圣地亚哥分校 Jacobs 眼底中心、Shiley 眼科研究所)