Diversity in data is critical for the successful training of deep learning models. Leveraged by a recurrent generative adversarial network, we propose the CT-SGAN model that generates large-scale 3D synthetic CT-scan volumes ($\geq 224\times224\times224$) when trained on a small dataset of chest CT-scans. CT-SGAN offers an attractive solution to two major challenges facing machine learning in medical imaging: a small number of given i.i.d. training data, and the restrictions around the sharing of patient data preventing to rapidly obtain larger and more diverse datasets. We evaluate the fidelity of the generated images qualitatively and quantitatively using various metrics including Fréchet Inception Distance and Inception Score. We further show that CT-SGAN can significantly improve lung nodule detection accuracy by pre-training a classifier on a vast amount of synthetic data.
Inter-vendor harmonization of Computed Tomography (CT) reconstruction kernels using unpaired image translation
Vanderbilt University(范德堡大学);Vanderbilt University Medical Center(范德堡大学医学中心);Amsterdam University Medical Center(阿姆斯特丹大学医学中心);University of Amsterdam(阿姆斯特丹大学)
GAN-based disentanglement learning for chest X-ray rib suppression
Radboud University Medical Center(拉德堡德大学医学中心);Netherlands Cancer Institute (NKI)(荷兰癌症研究所);Z 2 Sky Technologies Inc.(Z 2 Sky 技术公司);Johns Hopkins University(约翰斯·霍普金斯大学);University of Science and Technology of China(中国科学技术大学);Institute of Computing Technology, CAS(中国科学院计算技术研究所)
Computed Tomography (CT)-derived Cardiovascular Flow Estimation Using Physics-Informed Neural Networks Improves with Sinogram-based Training: A Simulation Study
Dept. of Bioengineering, University of California San Diego(加州大学圣地亚哥分校生物工程系);Dept. of Computer Science Engineering, University of California San Diego(加州大学圣地亚哥分校计算机科学与工程系);Dept. of Mechanical Engineering, Univ of Washington(华盛顿大学机械工程系);Depts of Mechanical Engineering and Cardiology, Univ. of Washington(华盛顿大学机械工程与心内科系);Depts. of Bioengineering, Radiology, University of California San Diego(加州大学圣地亚哥分校生物工程与放射学系)
Overcoming Barriers to Data Sharing with Medical Image Generation: A Comprehensive Evaluation
ETH Zurich(苏黎世联邦理工学院);Max Planck Institute for Intelligent Systems(马克斯·普朗克智能系统研究所);University Hospital of Tübingen(蒂宾根大学医院);Kantonsspital Winterthur(温特图尔州立医院);GlaxoSmithKline(葛兰素史克)
GAN-based Super-Resolution and Segmentation of Retinal Layers in Optical coherence tomography Scans
West Virginia University(西弗吉尼亚大学);Rockefeller Neuroscience Institute(洛克菲勒神经科学研究所);Georgia Institute of Technology(佐治亚理工学院);Twentieth Century Fox(二十世纪福克斯公司);Starfleet Academy(星际舰队学院)
Conditional GAN for Prediction of Glaucoma Progression with Macular Optical Coherence Tomography
Imperial College London(帝国理工学院);UCLA(加州大学洛杉矶分校);Jules Stein Eye Institute(朱尔斯·斯坦眼科研究所);California State University(加州州立大学);David Geffen School of Medicine(大卫·格芬医学院);Technische Universität München(慕尼黑工业大学)
BMD-GAN: Bone mineral density estimation using x-ray image decomposition into projections of bone-segmented quantitative computed tomography using hierarchical learning
Nara Institute of Science and Technology(奈良先端科学技术大学院大学);Osaka University Graduate School of Medicine(大阪大学大学院医学系研究科);Ehime University Graduate School of Medicine(爱媛大学大学院医学系研究科)
GAN-Based Architecture for Low-dose Computed Tomography Imaging Denoising
College of Life Science and Technology, Huazhong University of Science and Technology(华中科技大学生命科学技术学院);New York University Shanghai(上海纽约大学);School of Life Sciences and Biotechnology, Shanghai Jiao Tong University(上海交通大学生命科学技术学院)
OCT-GAN: Single Step Shadow and Noise Removal from Optical Coherence Tomography Images of the Human Optic Nerve Head
Singapore Eye Research Institute(新加坡眼科研究所);National University of Singapore(新加坡国立大学);Beihang University(北京航空航天大学);Duke-NUS Medical School(杜克-新加坡国立大学医学院)
American University of Beirut(贝鲁特美国大学);Reinstein Vision;London Vision Clinic(伦敦视力诊所);Columbia University Medical Center(哥伦比亚大学医学中心);Sorbonne Université(索邦大学);Biomedical Science Research Institute, Ulster University(阿尔斯特大学生物医学科学研究所);Bascom Palmer Eye Institute, University of Miami Miller School of Medicine(迈阿密大学米勒医学院巴斯康帕尔默眼科研究所);Stanford University(斯坦福大学);Saint Joseph University(圣约瑟夫大学);American University of Beirut Medical Center(贝鲁特美国大学医学中心)
Comparative clinical evaluation of "memory-efficient" synthetic 3d generative adversarial networks (gan) head-to-head to state of art: results on computed tomography of the chest
Politecnico di Milano(米兰理工大学);Fondazione IRCCS Policlinico San Matteo Pavia(圣马泰奥基金会IRCCS帕维亚医院);Università degli studi di Pavia(帕维亚大学);IULM University(IULM大学)
Department of Electrical and Computer Engineering, Sungkyunkwan University, Suwon, Korea(延世大学电气与计算机工程系);Department of Radiology and Center for Imaging Science, Samsung Medical Center, Sungkyunkwan University School of Medicine, Suwon, Korea(延世大学医学院放射科与成像科学中心)
Counterfactuals and Uncertainty-Based Explainable Paradigm for the Automated Detection and Segmentation of Renal Cysts in Computed Tomography Images: A Multi-Center Study
Maastricht University(马斯特里赫特大学);Memorial Sloan Kettering Cancer Center(纪念斯隆凯特琳癌症中心);Maastricht University Medical Center+(马斯特里赫特大学医学中心+);University of California-San Francisco(加州大学旧金山分校);University of Amsterdam(阿姆斯特丹大学);Amsterdam UMC(阿姆斯特丹大学医学中心);GROW Research Institute for Oncology and Reproduction(GROW肿瘤学与生殖学研究所)
Synthesizing Diverse Lung Nodules Wherever Massively: 3D Multi-Conditional GAN-based CT Image Augmentation for Object Detection
The University of Tokyo(东京大学);University of Cambridge(剑桥大学);Fujifilm Corporation(富士胶片公司);Jikei University School of Medicine(慈惠会医科大学);Juntendo University School of Medicine(顺天堂大学医学部)
精确的计算机辅助诊断依赖于大规模注释的病理图像,可以减轻忽视诊断的风险。不幸的是,在医学成像中,大多数可用的数据集都很小/碎片化。为了解决这个问题,作为一种数据增强(DA)方法,3D条件生成对抗网络(GANs)可以合成所需的逼真/多样化的3D图像作为额外的训练数据。然而,不存在用于一般的基于限界框的3D目标检测的基于3D条件GAN的DA方法,而与严格的3D分割不同,它可以以医生的最小注释成本定位疾病区域。此外,由于病变的位置/大小/衰减不同,因此进一步基于GAN的DA性能需要多种条件。因此,我们提出了3D多条件GAN(MC GAN)来生成自然放置在肺部计算机断层扫描图像上的逼真/多样化的32 X 32 X 32结节,以提高3D目标检测的灵敏度。我们的MCGAN采用两个判别器进行调节:上下文判别器学习用以噪声框为中心的环境对真实与合成结节/周围配对进行分类;结节判别器尝试用大小/衰减条件对真实与合成结节进行分类。结果表明,基于3D卷积神经网络的结节检测方法在固定的假阳性率下,可以在任何结节大小/衰减下实现更高的灵敏度,并克服了MCGAN生成的真实结节的医学数据不足--即使是专家医生也无法在视觉图灵测试中将其与真实结节区分开来。
A Data Augmentation Pipeline to Generate Synthetic Labeled Datasets of 3D Echocardiography Images using a GAN
GE Vingmed Ultrasound - GE Healthcare(GE Vingmed超声 - GE医疗);CARIM School for Cardiovascular Diseases(CARIM心血管疾病学院);Maastricht University Medical Center(马斯特里赫特大学医学中心)
Sparse-GAN: Sparsity-constrained Generative Adversarial Network for Anomaly Detection in Retinal OCT Image
ShanghaiTech University(上海科技大学);Cixi Institute of Biomedical Engineering, Chinese Academy of Sciences(中国科学院慈溪生物医学工程研究所);UBTech Research(优必选研究院);Southern University of Science and Technology(南方科技大学);Inception Institute of Artificial Intelligence(起源人工智能研究院)
HealthiVert-GAN: A Novel Framework of Pseudo-Healthy Vertebral Image Synthesis for Interpretable Compression Fracture Grading
School of Biomedical Engineering, Shanghai Jiao Tong University(上海交通大学生物医学工程学院);National Engineering Research Center of Advanced Magnetic Resonance Technologies for Diagnosis and Therapy (NERC-AMRT)(先进磁共振诊断与治疗技术国家工程研究中心);Med-X Research Institute, Shanghai Jiao Tong University(医-X研究院);Kashi Prefecture Second People’s Hospital(喀什县第二人民医院);Renji Hospital, Shanghai, China(仁济医院);Shanghai Sixth People’s Hospital, Shanghai, China(上海第六人民医院)
Noise Conscious Training of Non Local Neural Network powered by Self Attentive Spectral Normalized Markovian Patch GAN for Low Dose CT Denoising
在医疗实践中使用计算机断层扫描(CT)成像的爆炸性增长,提高了公众对患者相关辐射剂量的关注。然而,减少辐射剂量导致噪声和伪影增加,这不利地降低了扫描的可解释性。因此,一种先进的图像重建算法来提高低剂量ct的诊断性能成为研究人员的主要关注点,但由于该问题的不适定性,这具有挑战性。近年来,基于深度学习的技术已经成为低剂量CT(LDCT)去噪的主要方法。然而,仍然存在一些常见的瓶颈,这阻碍了基于深度学习的技术提供最佳性能。在这项研究中,我们试图减轻这些问题与三个新的吸积。首先,我们提出了一种新的卷积模块作为第一次尝试利用CT图像的邻域相似性去噪任务。我们提出的模块有助于提高去噪的显著幅度。接下来,我们转向CT噪声的非平稳性问题,并引入了一种新的噪声感知均方误差损失用于LDCT去噪。此外,上述损失也有助于减轻使用图像补丁训练CT去噪网络时所需的繁重工作。最后,我们提出了一种新的CT去噪任务的函数。传统的香草型协议往往忽略了精细的结构细节,而专注于全球协议。我们提出的神经网络利用自注意力和像素级GANs来恢复LDCT图像的诊断质量。我们的方法在2016年NIH-AAPM-Mayo Clinic低剂量CT Grand Challenge的公开数据集上进行了验证,其表现明显优于现有的最先进方法。
4D VQ-GAN: Synthesising Medical Scans at Any Time Point for Personalised Disease Progression Modelling of Idiopathic Pulmonary Fibrosis
Hawkes Institute, University College London(伦敦大学学院霍克斯研究所);Department of Computer Science, University College London(伦敦大学学院计算机科学系);Department of Med. Physics & Biomedical Engineering, University College London(伦敦大学学院医学物理与生物医学工程系);Department of Respiratory Medicine, University Hospitals Leuven(鲁汶大学医院呼吸医学科);NIHR Southampton Biomedical Research Centre and Clinical and Experimental Sciences, University of Southampton(南安普顿大学NIHR南安普顿生物医学研究中心与临床与实验科学系);UCL Respiratory, University College London(伦敦大学学院呼吸科)
Uncertainty-Aware Semi-Supervised Method Using Large Unlabeled and Limited Labeled COVID-19 Data
Deakin University(迪肯大学);Isfahan University of Technology(伊斯法罕理工大学);University of Birjand(比尔詹德大学);Ferdowsi University of Mashhad(马什哈德菲尔多西大学);K. N. Toosi University of Technology(图西理工大学);Universidad de Granada(格拉纳达大学);Dibrugarh University(迪布鲁格尔大学);Iran University of Medical Sciences(伊朗医科大学);The George Institute for Global Health(乔治全球健康研究院);University of Sydney(悉尼大学);Ngee Ann Polytechnic(义安理工学院);Singapore University of Social Sciences(新加坡新跃社科大学);Asia University(亚洲大学)
Photoacoustic image synthesis with generative adversarial networks
German Cancer Research Center (DKFZ)(德国癌症研究中心);Heidelberg University(海德堡大学);Helmholtz Information and Data Science School for Health(亥姆霍兹健康信息与数据科学学院);Cancer Research UK Cambridge Institute, University of Cambridge(剑桥大学英国癌症研究中心剑桥研究所)