Accurately segmenting fluid in 3D optical coherence tomography (OCT) images is critical for detecting eye diseases but remains challenging. Traditional autoencoder-based methods struggle with resolution loss and information recovery. While transformer-based models improve segmentation, they arent optimized for 3D OCT volumes, which vary by vendor and extraction technique. To address this, we propose SwinVFTR, a transformer architecture for precise fluid segmentation in 3D OCT images. SwinVFTR employs channel-wise volumetric sampling and a shifted window transformer block to improve fluid localization. Moreover, a novel volumetric attention block enhances spatial and depth-wise attention. Trained using multi-class dice loss, SwinVFTR outperforms existing models on Spectralis, Cirrus, and Topcon OCT datasets, achieving mean dice scores of 0.72, 0.59, and 0.68, respectively, along with superior performance in mean intersection-over-union (IOU) and structural similarity (SSIM) metrics.
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)
Denoising, segmentation and volumetric rendering of optical coherence tomography angiography (OCTA) image using deep learning techniques: a review
Department of Mechanical and Electronic Engineering, Kunming University of Science and Technology(昆明理工大学机械电子工程系);Eye Clinics, Yunnan First People's Hospital(云南第一人民医院眼科)
RetiNet: Automatic AMD identification in OCT volumetric data
University of Bern(伯尔尼大学);Lausanne University(洛桑大学);University Hospital, Lausanne(洛桑大学医院);Ecole Polytechnique Federale de Lausanne(洛桑联邦理工学院);Bern University Hospital(伯尔尼大学医院)
Temporal and volumetric denoising via quantile sparse image prior
Friedrich-Alexander-Universität Erlangen-Nürnberg(弗里德里希-亚历山大大学埃尔朗根-纽伦堡分校);Universitätsklinikum Erlangen(埃尔朗根大学医院);Massachusetts Institute of Technology(麻省理工学院);e.solutions GmbH(e.solutions有限公司)
VAMOS-OCTA: Vessel-Aware Multi-Axis Orthogonal Supervision for Inpainting Motion-Corrupted OCT Angiography Volumes
Department of Computer Science, Vanderbilt University(计算机科学系,范德比尔特大学);Digital Technology and Innovation, Siemens Healthineers(数字技术与创新,西门子医疗);Department of Biomedical Engineering, Vanderbilt University(生物医学工程系,范德比尔特大学)
PC-Seg: Progressive Cross-View Consistency for 3D OCT Segmentation from Sparse 2D Annotations
Graduate School of Information Sciences, Tohoku University(东北大学信息科学研究生院);Department of Ophthalmology, Graduate School of Medicine, Tohoku University(东北大学医学研究生院眼科系);Institute of Ophthalmology, University College London(伦敦大学学院眼科研究所);NIHR Biomedical Research Centre at Moorfields Eye Hospital NHS Foundation Trust(摩尔菲尔德眼科医院NHS基金会信托基金NIHR生物医学研究中心);UCL Hawkes Institute, University College London(伦敦大学学院UCL霍克斯研究所);Department of Computer Science, University College London(伦敦大学学院计算机科学系)
Efficient and high accuracy 3-D OCT angiography motion correction in pathology
Friedrich-Alexander-Universität Erlangen-Nürnberg(弗里德里希-亚历山大大学埃尔朗根-纽伦堡分校);Massachusetts Institute of Technology(麻省理工学院);New England Eye Center(新英格兰眼科中心);Tufts Medical Center(塔夫茨医疗中心)
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)
Assignment Flow for Order-Constrained OCT Segmentation
Heidelberg University(海德堡大学)
目前,光学相干断层扫描(OCT)是最常用的非侵入性成像方法之一,用于采集人体视网膜组织和血管的大体积扫描。为了从提取的OCT体积中解析决定性信息并使其适用于进一步的诊断分析,视网膜层厚度的准确识别是为每个患者单独完成的重要任务。然而,连续多次OCT扫描的手动检查是一项要求很高且耗时的任务,这导致了冗长的鉴定过程,并且在存在组织相关散斑噪声的情况下经常混淆。因此,自动分割模型的阐述已成为医学图像处理领域的一项重要任务。我们提出了一种新的,纯数据驱动的\textit{geometric approach to order-constrained 3D OCT retinal cell layer segmentation},它需要作为输入数据在任何度量空间,并伴随着基本的操作,可以有效地并行计算。相对于许多已建立的视网膜检测方法,我们提出的配方避免了使用任何形状之前,并完成了视网膜的自然顺序在一个纯粹的几何方式。这使得该方法是无偏的,因此适合于检测视网膜组织结构的局部解剖变化。为了证明所提出的方法的鲁棒性,我们比较了两种不同的选择功能的数据集上的手动注释的3D OCT卷的健康人的视网膜。计算分割的质量进行比较,以平均绝对误差和骰子相似系数方面的最先进的。结果表明,一个巨大的潜力,将我们的方法应用于疾病视网膜的分类,并打开了一个新的研究方向,关于联合分割视网膜细胞层和血管结构。
Neural network-based image reconstruction in swept-source optical coherence tomography using undersampled spectral data
University of California, Los Angeles(加利福尼亚大学洛杉矶分校);California NanoSystems Institute(加州纳米系统研究所);University of Houston(休斯顿大学);Baylor College of Medicine(贝勒医学院)
3DTINC: Time-Equivariant Non-Contrastive Learning for Predicting Disease Progression from Longitudinal OCTs
Medical University of Vienna(维也纳医科大学);BioMedIA, Imperial College London(伦敦帝国理工学院生物医学图像分析研究所);Institute for AI and Informatics in Medicine, Klinikum rechts der Isar, Technical University Munich(慕尼黑工业大学伊萨尔河畔医院医学人工智能与信息学研究所);Department of Ophthalmology, University of Basel(巴塞尔大学眼科系)
Pretrained Deep 2.5D Models for Efficient Predictive Modeling from Retinal OCT
University of Basel(巴塞尔大学);Technical University Munich(慕尼黑工业大学);Medical University of Vienna(维也纳医科大学);Moorfields Eye Hospital NHS Foundation Trust(摩尔菲尔斯眼科医院NHS基金会信托);Imperial College London(帝国理工学院);Institute of Molecular and Clinical Ophthalmology Basel(巴塞尔分子与临床眼科研究所);University of Southampton(南安普顿大学)
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)
APTOS-2024 challenge report: Generation of synthetic 3D OCT images from fundus photographs
The Hong Kong Polytechnic University(香港理工大学);Rangsit University(兰实大学);Rajavithi Hospital(拉差维提医院);Monash University(莫纳什大学);Nanjing University of Science and Technology(南京理工大学);Institute of High Performance Computing (IHPC), Agency for Science, Technology and Research (A*STAR)(新加坡科技研究局高性能计算研究院);Mayo Clinic(梅奥诊所);China Medical University(中国医科大学);Northeastern University(东北大学);Henan Kaifeng College of Science Technology and Communication(河南开封科技传媒学院);University of Electronic Science and Technology of China(电子科技大学);Centre for Eye and Vision Research (CEVR)(眼科与视觉研究中心)