Simulation of Brain Resection for Cavity Segmentation Using Self-Supervised and Semi-Supervised Learning
Wellcome EPSRC Centre for Interventional and Surgical Sciences (WEISS), University College London（伦敦大学学院威尔康姆EPSRC介入与外科科学中心（WEISS））；School of Biomedical Engineering and Imaging Sciences (BMEIS), King’s College London（伦敦国王学院生物医学工程与成像科学学院（BMEIS））；Department of Clinical and Experimental Epilepsy, UCL Queen Square Institute of Neurology, London（伦敦大学学院女王广场神经病学研究所临床与实验癫痫系）；National Hospital for Neurology and Neurosurgery, Queen Square, London（伦敦女王广场神经与神经外科医院）
Fernando Pérez-García, Roman Rodionov, Ali Alim-Marvasti, Rachel Sparks, John S. Duncan, Sébastien Ourselin
备注：
13 pages, 6 figures, accepted at the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) 2020
发表信息：
Medical Image Computing and Computer Assisted Intervention - MICCAI 2020. Lecture Notes in Computer Science, vol 12263. Springer, Cham
Resective surgery may be curative for drug-resistant focal epilepsy, but only 40% to 70% of patients achieve seizure freedom after surgery. Retrospective quantitative analysis could elucidate patterns in resected structures and patient outcomes to improve resective surgery. However, the resection cavity must first be segmented on the postoperative MR image. Convolutional neural networks (CNNs) are the state-of-the-art image segmentation technique, but require large amounts of annotated data for training. Annotation of medical images is a time-consuming process requiring highly-trained raters, and often suffering from high inter-rater variability. Self-supervised learning can be used to generate training instances from unlabeled data. We developed an algorithm to simulate resections on preoperative MR images. We curated a new dataset, EPISURG, comprising 431 postoperative and 269 preoperative MR images from 431 patients who underwent resective surgery. In addition to EPISURG, we used three public datasets comprising 1813 preoperative MR images for training. We trained a 3D CNN on artificially resected images created on the fly during training, using images from 1) EPISURG, 2) public datasets and 3) both. To evaluate trained models, we calculate Dice score (DSC) between model segmentations and 200 manual annotations performed by three human raters. The model trained on data with manual annotations obtained a median (interquartile range) DSC of 65.3 (30.6). The DSC of our best-performing model, trained with no manual annotations, is 81.7 (14.2). For comparison, inter-rater agreement between human annotators was 84.0 (9.9). We demonstrate a training method for CNNs using simulated resection cavities that can accurately segment real resection cavities, without manual annotations.
Accurate segmentation of brain resection cavities (RCs) aids in postoperative analysis and determining follow-up treatment. Convolutional neural networks (CNNs) are the state-of-the-art image segmentation technique, but require large annotated datasets for training. Annotation of 3D medical images is time-consuming, requires highly-trained raters, and may suffer from high inter-rater variability. Self-supervised learning strategies can leverage unlabeled data for training. We developed an algorithm to simulate resections from preoperative magnetic resonance images (MRIs). We performed self-sup…
Malignant brain tumors are among the deadliest neoplasms with the lowest survival rates of any cancer type. In considering surgical tumor resection, suboptimal extent of resection is linked to poor clinical outcomes and lower overall survival rates. Currently available tools for intraoperative histopathological assessment require an average of 20 minutes processing and are of limited diagnostic quality for guiding surgeries. Consequently, there is an unaddressed need for a rapid imaging technique to guide maximal resection of brain tumors. Working towards this goal, presented here is an all op…
Intraoperative Glioma Segmentation with YOLO + SAM for Improved Accuracy in Tumor Resection
Carnegie Vanguard High School（卡内基先锋高中）；Devon Preparatory High School（德文预备高中）；Lake Travis High School（特拉维斯湖高中）；Irvington High School（欧文顿高中）；Algoverse AI Research（Algoverse AI研究机构）
Towards Machine Learning-based Quantitative Hyperspectral Image Guidance for Brain Tumor Resection
University of British Columbia（不列颠哥伦比亚大学）；University Hospital of Münster（明斯特大学医院）；Macquarie University（麦考瑞大学）；Hokkaido Medical Center, National Hospital Organization（国立医院机构北海道医疗中心）
University of Bern（伯尔尼大学）；University Hospital and University of Bern（伯尔尼大学医院和伯尔尼大学）
This paper proposes a novel approach for uncertainty quantification in dense Conditional Random Fields (CRFs). The presented approach, called Perturb-and-MPM, enables efficient, approximate sampling from dense multi-label CRFs via random perturbations. An analytic error analysis was performed which identified the main cause of approximation error as well as showed that the error is bounded. Spatial uncertainty maps can be derived from the Perturb-and-MPM model, which can be used to visualize uncertainty in image segmentation results. The method is validated on synthetic and clinical Magnetic R…
Spatio-spectral classification of hyperspectral images for brain cancer detection during surgical operations
University of Las Palmas de Gran Canaria（大加那利岛拉斯帕尔马斯大学）；Imperial College London（伦敦帝国理工学院）；University Hospital Doctor Negrin（内格林大学医院）；University Hospital Southampton（南安普顿大学医院）；University of Cambridge（剑桥大学）；Universidad Politecnica de Madrid（马德里理工大学）
The paper demonstrates the use of the fully convolutional neural network for glioma segmentation on the BraTS 2019 dataset. Three-layers deep encoder-decoder architecture is used along with dense connection at encoder part to propagate the information from coarse layer to deep layers. This architecture is used to train three tumor sub-components separately. Subcomponent training weights are initialized with whole tumor weights to get the localization of the tumor within the brain. At the end, three segmentation results were merged to get the entire tumor segmentation. Dice Similarity of traini…
The Hong Kong University of Science and Technology（香港科技大学）
Registration of pre-operative and post-recurrence brain images is often needed to evaluate the effectiveness of brain gliomas treatment. While recent deep learning-based deformable registration methods have achieved remarkable success with healthy brain images, most of them would be unable to accurately align images with pathologies due to the absent correspondences in the reference image. In this paper, we propose a deep learning-based deformable registration method that jointly estimates regions with absent correspondence and bidirectional deformation fields. A forward-backward consistency c…
Data-driven registration and modeling of brain deformation for image-guided neurosurgery
LASIGE, Faculty of Sciences, University of Lisbon（里斯本大学科学学院LASIGE）；Department of Neurosurgery and Department of Radiology, Brigham and Women's Hospital, Harvard Medical School（哈佛医学院布里洛妇女医院神经外科与放射科）；Gina Cody School of Engineering and Computer Science, Concordia University（康科迪亚大学工程与计算机科学学院）；Cancer Research UK Cambridge Centre, University of Cambridge（剑桥大学癌症研究英国中心）；Department of Oncology, University of Cambridge（剑桥大学肿瘤科）；Department of Clinical Neurosciences, University of Cambridge（剑桥大学临床神经科学系）；Computational Health Informatics Program (CHIP) and Department of Radiology, Boston Children's Hospital, Harvard Medical School（哈佛医学院波士顿儿童医院计算健康信息学计划与放射科）；Sorbonne Université, Institut du Cerveau - Paris Brain Institute - ICM（索邦大学巴黎脑研究所-ICM）
Arizona State University（亚利桑那州立大学）；Barrow Neurological Institute（巴罗神经学研究所）；Irkutsk State Medical University（伊尔库茨克州立医科大学）
Confocal Laser Endomicroscope (CLE) is a novel handheld fluorescence imaging device that has shown promise for rapid intraoperative diagnosis of brain tumor tissue. Currently CLE is capable of image display only and lacks an automatic system to aid the surgeon in analyzing the images. The goal of this project was to develop a computer-aided diagnostic approach for CLE imaging of human glioma with feature localization function. Despite the tremendous progress in object detection and image segmentation methods in recent years, most of such methods require large annotated datasets for training. H…
Uncovering Neuroimaging Biomarkers of Brain Tumor Surgery with AI-Driven Methods
Department of Communication Engineering University of Málaga（通信工程系马德里大学）；Department of Clinical Neurosciences University of Cambridge（临床神经科学系剑桥大学）；Department of Neuroscience University of Padova（神经科学系帕多瓦大学）；Department of Signal Theory, Telematics and Communications, University of Granada（信号理论、电信与通信系格拉纳达大学）；Department of Computer Science and Technology University of Cambridge（计算机科学与技术系剑桥大学）；Department of Psychiatry University of Cambridge（精神病学系剑桥大学）；NHS Foundation Trust United Kingdom（国家健康服务基金会信托英国）
Automatic brain tumor segmentation in 2D intra-operative ultrasound images using magnetic resonance imaging tumor annotations
Department of Health Research（健康研究系）；SINTEF Digital（SINTEF数字）；Department of Physics（物理系）；Norwegian University of Science and Technology (NTNU)（挪威科学与技术大学（NTNU））；Department of Circulation and Medical Imaging（循环与医学成像系）；St. Olavs hospital（圣奥拉夫医院）；Department of Neuromedicine and Movement Science（神经医学与运动科学系）
Automatic methods for measuring normalized regional brain volumes from MRI data are a key tool to help in the objective diagnostic and follow-up of many neurological diseases. To estimate such regional brain volumes, the intracranial cavity volume is commonly used for normalization. In this paper, we present an accurate and efficient approach to automatically segment the intracranial cavity using a volumetric 3D convolutional neural network and a new 3D patch extraction strategy specially adapted to deal with the traditional low number of training cases available in supervised segmentation and…
Full-Head Segmentation of MRI with Abnormal Brain Anatomy: Model and Data Release
The City College of New York（纽约市立学院）；New York University（纽约大学）；Georgetown University（乔治城大学）；University of North Carolina School of Medicine（北卡罗来纳大学医学院）；Soterix Medical
