HistoTools: A suite of digital pathology tools for quality control, annotation and dataset identification

HistoTools：一套用于质量控制、注释和数据集识别的数字病理学工具

• 批准号: 9897498
• 负责人: Andrew Robert Janowczyk
• 金额: $ 38.11万
• 依托单位: CASE WESTERN RESERVE UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-03-20 至 2022-02-28
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/9897498
• 关键词: Active Learning; Address; Adoption; Algorithms; American; American Society of Clinical Oncology; Automobile Driving; Cancer Patient; Cell Nucleus; Characteristics; Classification; Clinical; Clinical Pathology; Clinical Research; Clinical Trials; Collection; Communities; Computer Assisted; Data; Data Set; Detection; Development; Diagnosis; Eastern Cooperative Oncology Group; Employment; Ensure; Environment; Estrogen receptor positive; Evaluation; Feedback; Generations; Histologic; Histology; Image; Immunotherapy; International; Label; Learning; Lymphocyte; Machine Learning; Malignant Neoplasms; Malignant neoplasm of lung; Manuals; Masks; Mitosis; Modeling; Morphologic artifacts; Morphology; Nature; Nivolumab; Noise; Non-Small-Cell Lung Carcinoma; Nuclear; Oncology; Optics; Outcome; Paper; Pathologist; Patients; Performance; Population; Preparation; Process; Quality Control; Recurrence; Reproducibility; Research; Role; Scanning; Slide; Societies; Stains; Technology; Testing; The Cancer Imaging Archive; Time; Tissues; Training; Tumor-Infiltrating Lymphocytes; Validation; Visualization; Weight; Work; anticancer research; base; cancer diagnosis; cancer recurrence; cohort; combat; companion diagnostics; deep learning; design; diagnostic assay; digital; digital pathology; experimental study; heterogenous data; high resolution imaging; imaging informatics; improved; indexing; industry partner; innovation; interactive tool; interest; large datasets; learning network; malignant breast neoplasm; open source; outcome forecast; outcome prediction; pathology imaging; photonics; predicting response; prognostic; prototype; quantitative imaging; repository; response; tool; tumor heterogeneity

ABSTRACT: Roughly 40% of the US population will be diagnosed with some form of cancer in their lifetime. In a majority of these cases, a definitive cancer diagnosis is only possible via histopathologic confirmation using a tissue slide. Increasingly, these slides are being digitally scanned as high-resolution images for usage in both clinical and research digital pathology (DP) workflows. Our group has been pioneering the use of deep learning (DL), a form of machine learning, for segmentation, detection, and classification of various cancers using digital pathology images. DL learns features and their associated weighting from large datasets to maximally discriminate between user labeled data (e.g., cancer vs non-cancer, nuclei vs non-nuclei); a paradigm known as “learn from data”. Unfortunately, this paradigm makes DL especially sensitive to low quality slides, noise induced by small errors in the manual user labeling process, and general dataset heterogeneity. As many groups do not intentionally account for these problems, they learn that successful employment of DL technologies relies heavily on explicitly addressing challenges associated with (a) carefully curating high quality slides without preparation or scanning artifacts, (b) obtaining a large precise collection of annotations delineating objects of interest, and (c) selecting diverse datasets to ensure robust classifier performance when clinically deploying the model. To address these challenges we propose HistoTools, a suite of three modules or “Apps”: (1) HistoQC examines slides for artifacts and computes metrics associated with slide presentation characteristics (e.g., stain intensity, compression levels) helping to quantify ranges of acceptable characteristics for downstream algorithmic evaluation. (2) HistoAnno drastically improves the efficiency of annotation efforts using a combined active learning and deep learning approach to ensure experts focus only on regions which are important for classifier improvement. (3) HistoFinder aids in selecting suitable training and test cohorts to guarantee that various tissue level characteristics are well balanced, leading to increased reproducibility. Our team already has working prototypes of HistoQC (100% concordance with a pathologist, evaluated on n>1200 slides) and HistoAnno (30% efficiency improvement during annotation tasks). In this U01, we seek to further develop and evaluate HistoTools in the context of enhancing two companion diagnostic (CDx) assays being developed in our group. First, we will use HistoTools to quality control and annotate nuclei, tubules, and mitosis for improving our CDx classifier for predicting recurrence in breast cancers using a cohort of n>900 patients from completed trial ECOG 2197. Secondly, HistoTools will be employed for quality control and identification of tumor infiltrating lymphocytes and cancer nuclei towards improving our CDx classifier for predicting response to immunotherapy in lung cancer using the n>700 patients from completed clinical trials Checkmate 017 and 057. These tools will build on our existing open source tool repository to aid in real-time feedback and dissemination throughout the ITCR and cancer research community.

摘要：大约40%的美国人口在其一生中会被诊断出患有某种形式的癌症。在 在大多数这些病例中，只有通过组织病理学确认，使用 组织切片这些幻灯片越来越多地被数字扫描成高分辨率图像， 临床和研究数字病理学（DP）工作流程。我们的团队一直是深度学习的先驱 (DL)这是一种机器学习形式，用于使用数字技术对各种癌症进行分割、检测和分类。 病理图像DL从大型数据集中学习特征及其相关权重， 区分用户标记的数据（例如，癌症与非癌症，细胞核与非细胞核）;已知的范例 “从数据中学习”。不幸的是，这种模式使得DL对低质量的幻灯片、噪音和其他干扰特别敏感。 由手动用户标记过程中的小错误和一般数据集异质性引起。一样多 团体不故意占这些问题，他们了解到，成功就业的DL 技术在很大程度上依赖于明确解决与以下方面相关的挑战：（a）精心策划高 没有制备或扫描伪影的高质量载玻片，（B）获得大量精确的注释集合 描绘感兴趣的对象，以及（c）选择不同的数据集，以确保在以下情况下的鲁棒分类器性能： 在临床上部署模型。为了应对这些挑战，我们提出了HistoTools，一套三个模块 或“应用程序”：（1）HistoQC检查幻灯片的伪影，并计算与幻灯片演示相关的指标 特性（例如，染色强度、压缩水平），有助于量化可接受的 下游算法评估的特征。(2)HistoAnno大大提高了 使用结合主动学习和深度学习方法的注释工作，以确保专家只关注 对分类器改进很重要的区域。(3)组织学有助于选择适当的培训 和测试队列，以保证各种组织水平的特征得到很好的平衡，从而增加 再现性我们的团队已经有了HistoQC的工作原型（与病理学家100%一致， 在n>1200个载玻片上评估）和HistoAnno（注释任务期间效率提高30%）。在U 01中， 我们寻求进一步开发和评估HistoTools的背景下，加强两个伴随诊断 (CDx)我们小组正在开发的检测方法。首先，我们将使用HistoTools进行质量控制和注释细胞核， 小管和有丝分裂，以改善我们的CDx分类器，用于使用队列预测乳腺癌的复发 来自已完成试验ECOG 2197的n>900例患者。其次，HistoTools将用于质量控制 以及肿瘤浸润淋巴细胞和癌细胞核的鉴定，以改进我们的CDx分类器， 使用来自已完成临床试验的n>700例患者预测肺癌对免疫疗法的反应 将军017和057这些工具将建立在我们现有的开源工具存储库之上，以提供实时帮助 在整个ITCR和癌症研究界的反馈和传播。