Machine learning algorithms to analyze large medical image datasets

用于分析大型医学图像数据集的机器学习算法

• 批准号: 10584569
• 负责人: SIMON K WARFIELD
• 金额: $ 37.61万
• 依托单位: BOSTON CHILDREN'S HOSPITAL
• 依托单位国家: 美国
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-07-01 至 2025-03-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10584569
• 关键词: Active Learning; Address; Adoption; Algorithms; Autism Diagnosis; Benchmarking; Brain; Characteristics; Child; Clinical; Collection; Crowding; Data; Data Collection; Data Set; Databases; Development; Fatigue; Image; Image Analysis; Informatics; Interruption; Label; Learning; Life; Machine Learning; Medical Imaging; Methods; Modeling; Morphologic artifacts; Noise; Patient Care; Patient-Focused Outcomes; Patients; Performance; Reference Standards; Research; Sampling; Speed; Structure; Techniques; Training; Uncertainty; Update; autism spectrum disorder; clinical decision-making; cost; crowdsourcing; deep learning; design; experimental study; imaging Segmentation; imaging biomarker; improved; innovation; insight; learning algorithm; learning strategy; machine learning algorithm; novel; novel strategies; prevent; quantitative imaging; radiologist; success; supervised learning

Machine learning (ML) is poised to enable faster and more accurate interpretation of medical images by augmenting the capabilities of experts. The cost and difficulty of generating expert quality labelled image data is the primary limitation preventing faster progress and deployment in more domains. Success of ML techniques for medical image interpretation may reduce the burden on radiologists, reducing errors arising from fatigue or interruption, while simultaneously reducing costs and increasing speed and accuracy for patients. Our overall objective for this research is to dramatically reduce the burden of creating high quality reference labels by requiring only a small set of such labels from experts. We propose to address this problem by creating innovative algorithms that will construct reference quality labelled data with little input from domain experts, thus dramatically reducing the cost of labelling. This will enable us to apply ML techniques to generate high quality labels of the large amounts of unlabeled data that are already available, which in turn will facilitate the assessment of potential quantitative imaging biomarkers. We will develop, extend and evaluate novel algorithms that represent three distinct strategies for reducing labelling cost. These three strategies are learning from unlabelled data incorporating a novel strategy for characterizing uncertainty, optimizing sample selection for expert quality labelling with a novel form of Active Learning especially suited for deep learning, and reducing the cost of achieving quality labeling by replacing or augmenting an expert with a crowd of inexperts. We will then implement and distribute these novel algorithms, facilitating the replication of our experiments. Finally, we will demonstrate the practical efficacy of these three strategies by applying them to the important challenge of identifying quantitative imaging biomarkers that best capture alterations in brain structure that are associated with characteristics of ASD. These fundamental advances in informatics algorithms will reduce the cost and increase the rate of obtaining quality labels, which will in turn facilitate the widespread adoption and deployment of machine learning algorithms for image interpretation. Ultimately, this will stimulate the development of new imaging biomarkers that hold the potential to dramatically improve clinical decision-making and patient outcomes.

机器学习（ML）有望通过以下方式更快、更准确地解释医学图像： 增强专家的能力。生成专家质量标记图像数据的成本和困难 是阻止在更多域中更快地进行开发和部署的主要限制。ML的成功 用于医学图像解释的技术可以减轻放射科医师的负担， 减少疲劳或中断，同时降低成本并提高速度和准确性， 患者我们这项研究的总体目标是大大减轻创造高质量产品的负担。 通过仅要求来自专家的一小组这样的标签来参考标签。我们建议解决这个问题 通过创建创新的算法，将构建参考质量标记的数据， 专家，从而大大降低了标签的成本。这将使我们能够应用ML技术来生成 高质量的标签大量的未标记的数据，这反过来又将促进 潜在的定量成像生物标志物的评估。我们将开发，扩展和评估新的 这些算法代表了降低标签成本的三种不同策略。这三个战略是 从未标记的数据中学习，并采用新的策略来表征不确定性，优化样本 选择专家质量标签，采用特别适合深度学习的新型主动学习， 并通过用一群专家来取代或增加专家来降低实现质量标签的成本， 不专家。然后，我们将实现和分发这些新的算法，促进我们的复制。 实验最后，我们将通过将这三种策略应用于 识别最能捕捉大脑变化的定量成像生物标志物的重要挑战 与ASD特征相关的结构。信息学的这些基本进步 算法将降低成本并提高获得质量标签的比率，这反过来又将促进 广泛采用和部署机器学习算法进行图像解释。最终这 将刺激新的成像生物标志物的发展，这些生物标志物有可能显着改善 临床决策和患者结局。