On the Effect of Inter-observer Variability for a Reliable Estimation of Uncertainty of Medical Image Segmentation
Institute for Surgical Technologies and Biomechanics, University of Bern(伯尔尼大学外科技术与生物力学研究所);SCAN, Institute for Diagnostic and Interventional Neuroradiology(诊断与介入神经放射学研究所SCAN);University Clinic for Radio-oncology, Inselspital, Bern University Hospital, University of Bern(伯尔尼大学因塞尔医院放射肿瘤学大学诊所)
Alain Jungo, Raphael Meier, Ekin Ermis, Marcela Blatti-Moreno, Evelyn Herrmann, Roland Wiest, Mauricio Reyes
备注:
Appears in Medical Image Computing and Computer Assisted Interventions (MICCAI), 2018
Uncertainty estimation methods are expected to improve the understanding and quality of computer-assisted methods used in medical applications (e.g., neurosurgical interventions, radiotherapy planning), where automated medical image segmentation is crucial. In supervised machine learning, a common practice to generate ground truth label data is to merge observer annotations. However, as many medical image tasks show a high inter-observer variability resulting from factors such as image quality, different levels of user expertise and domain knowledge, little is known as to how inter-observer variability and commonly used fusion methods affect the estimation of uncertainty of automated image segmentation. In this paper we analyze the effect of common image label fusion techniques on uncertainty estimation, and propose to learn the uncertainty among observers. The results highlight the negative effect of fusion methods applied in deep learning, to obtain reliable estimates of segmentation uncertainty. Additionally, we show that the learned observers' uncertainty can be combined with current standard Monte Carlo dropout Bayesian neural networks to characterize uncertainty of model's parameters.
Human-level Performance On Automatic Head Biometrics In Fetal Ultrasound Using Fully Convolutional Neural Networks
Imperial College London(帝国理工学院);ETH Zurich(苏黎世联邦理工学院);Guy’s and St Thomas’ NHS Foundation(盖伊和圣托马斯国民保健服务基金会)
从胎儿超声图像中测量头部生物特征对于监测胎儿的健康发育具有重要意义。然而,相关解剖结构的准确测量在临床中受到较大的观察者间差异的影响。为了解决这个问题,提出了一种利用全卷积网络(FCN)来确定胎儿头围(HC)和双顶径(BPD)测量值的自动化方法。在常规筛查检查期间,在头部的大约2000个2D超声图像上训练FCN,所述图像具有由45个不同的超声医师提供的注释,以执行头部的语义分割。椭圆被拟合到所得到的分割轮廓,以模仿通常由声谱仪操作员产生的注释。该模型的性能与观察者间的变异性进行了比较,其中两位专家手动注释了100张测试图像。HC的平均绝对模型专家误差略优于观察者间误差(1.99 mm vs 2.16 mm),BPD(0.61 mm vs 0.59 mm)以及Dice系数(0.980 vs 0.980)相当。我们的研究结果表明,该模型的性能与人类专家相似,并学会从由许多超声医师注释的大型数据集中产生准确的预测。此外,在GPU上以15 fps的速度近实时生成测量结果,这可以加快熟练和实习超声医师的临床工作流程。
Deep Learning-Based Quantification of Pulmonary Hemosiderophages in Cytology Slides
Friedrich-Alexander-Universität Erlangen-Nürnberg(埃尔朗根-纽伦堡弗里德里希-亚历山大大学);EUROIMMUN Medizinische Labordiagnostika AG(欧蒙医学实验诊断股份公司);Freie Universität Berlin(柏林自由大学);VetPath Laboratory Services(VetPath实验室服务机构);Laboklin GmbH und Co. KG(拉博克林有限两合公司);Universitätsklinikum Erlangen(埃尔朗根大学医院);Friedrich Alexander University Erlangen-Nürnberg(埃尔朗根-纽伦堡弗里德里希-亚历山大大学);Equine Clinic, Freie Universität Berlin(柏林自由大学马科动物诊所);Institute of Neuropathology, Friedrich Alexander University Erlangen-Nürnberg(埃尔朗根-纽伦堡弗里德里希-亚历山大大学神经病理学研究所)
How Many Annotators Do We Need? -- A Study on the Influence of Inter-Observer Variability on the Reliability of Automatic Mitotic Figure Assessment
Friedrich-Alexander-Universität Erlangen-Nürnberg(埃尔朗根-纽伦堡弗里德里希-亚历山大大学);Freie Universität Berlin(柏林自由大学);Animal Medical Center(动物医疗中心);University of Pennsylvania(宾夕法尼亚大学);University of Veterinary Medicine Hannover(汉诺威兽医大学);Synlab’s VPG Histology(Synlab VPG组织学部门);Michigan State University(密歇根州立大学);Idorsia Pharmaceuticals Ltd(Idorsia制药有限公司);International Zoo Veterinary Group(国际动物园兽医集团);University of Bern(伯尔尼大学);IDEXX Vet Med Labor GmbH(爱德士兽医实验室有限责任公司)
Label uncertainty-guided multi-stream model for disease screening
Zhongshan Ophthalmic Center, Sun Yat-sen University(中山大学中山眼科中心);Monash eResearch Centre, Monash University(蒙纳士大学电子研究中心);Centre for Eye Research, Melbourne University(墨尔本大学眼科研究中心)
Deep learning-based segmentation of T1 and T2 cardiac MRI maps for automated disease detection
Transilvania University of Brasov(布拉索夫特兰西瓦尼亚大学);Siemens SRL(西门子有限责任公司);Robert Bosch Hospital(罗伯特·博世医院);Siemens Healthcare GmbH(西门子医疗有限责任公司);Siemens Medical Solutions USA, Inc.(西门子医疗解决方案美国公司);Siemens Healthineers AG(西门子医疗股份公司)
A Deep Learning Localization Method for Measuring Abdominal Muscle Dimensions in Ultrasound Images
College of Science and Engineering, James Cook University(詹姆斯·库克大学理工学院);McGill University(麦吉尔大学);Element AI(Element AI公司);College of Healthcare Sciences, James Cook University(詹姆斯·库克大学健康科学学院)
Comprehensive framework for evaluation of deep neural networks in detection and quantification of lymphoma from PET/CT images: clinical insights, pitfalls, and observer agreement analyses
University of British Columbia(英属哥伦比亚大学);BC Cancer Research Institute(BC癌症研究所);Microsoft AI for Good Lab(微软AI公益实验室);Pomeranian Medical University(波美拉尼亚医科大学);BC Children’s Hospital(BC儿童医院);St. Mary’s Hospital(圣玛丽医院);BC Cancer(BC癌症中心)
Walkable to Whom? Capturing Subjective Variability in Walkability Perception Using Multimodal Deep Learning
School of Civil and Environmental Engineering, University of New South Wales (UNSW)(新南威尔士大学土木与环境工程学院);Research Centre for Integrated Transport Innovation (rCITI)(综合交通创新研究中心)