Rad-pathomic deep learning models to assist radiologists in differentiating aggressive from indolent prostate cancer on MRI

放射病理深度学习模型可帮助放射科医生在 MRI 上区分侵袭性前列腺癌和惰性前列腺癌

• 批准号: 10549719
• 负责人: Mirabela Rusu
• 金额: $ 53.61万
• 依托单位: STANFORD UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-01-12 至 2026-12-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10549719
• 关键词: 3-Dimensional; Affect; Agreement; American; Appearance; Atlas of Cancer Mortality in the United States; Benign; Biological Markers; Biopsy; Biopsy Specimen; Cancer Detection; Cessation of life; Data Set; Detection; Early Diagnosis; Hemorrhage; Indolent; Infection; Label; Learning; Lesion; Machine Learning; Magnetic Resonance Imaging; Malignant Neoplasms; Malignant neoplasm of prostate; Maps; Methods; Modeling; Pain; Pathologist; Pathology; Patients; Pattern; Positioning Attribute; Prostate; Prostatectomy; Radical Prostatectomy; Reader; Screening for Prostate Cancer; Slice; Specificity; Testing; Tissues; Training; Visual; cancer care; deep learning; deep learning model; detection method; improved; indexing; innovation; learning strategy; magnetic resonance imaging biomarker; men; multidisciplinary; novel; pathology imaging; prostate biopsy; radiological imaging; radiologist; side effect

PROJECT SUMMARY/ABSTRACT Prostate cancer is the second deadliest cancer for American men. MRI is increasingly used to guide prostate biopsies and has potential to spare 500,000 men/year from the side effects of invasive biopsies. Yet, subtle differences in MRI appearance of aggressive vs. indolent (non-lethal) cancer vs. benign tissue creates three problems: missed cancers, high rates of false positives, and only moderate inter-reader agreement among ra- diologists. Selective identification of aggressive and indolent cancers is imperative for reducing cancer death while minimizing side effects from unneeded biopsies. We propose to develop and use pathology-based (pathomic) MRI biomarkers in rad-pathomic deep learning methods to assist radiologists in detecting and localizing aggressive vs. indolent cancers on prostate MRI. In addition, our proposed method will be the first to localize aggressive and indolent cancers when they coexist (76% of index lesions). We performed four preliminary studies in our unique dataset of matched radiology and pathology images. First, we found a high agreement in labeling aggressive vs. indolent cancers between the automated method and two pathologists. Second, we developed pathomic MRI bi- omarkers from MRI features that correlate with features derived from pathology images. Third, we used the biomarkers in rad-pathomic deep learning models to detect cancer (AUC: 0.86) and aggressive cancer (AUC: 0.85) on MRI. Fourth, we showed that combining radiologists and the rad-pathomic deep learning models helped identify 14% more aggressive cancers missed by radiologists. Three innovations will improve the localization of aggressive vs. indolent cancers on prostate MRI. First, we will develop 3D RAPSODI, a novel 3D registration method for 3D reconstructed MRI and pathology images to eliminate the need for slice-to-slice correspondences and map cancer labels from pathology onto MRI. Second, we will leverage our correlation learning method to identify pathomic MRI biomarkers. Third, we will use deep learning models to assist radiologists in localizing aggressive cancer on MRI. Our multidisciplinary team is uniquely positioned to test whether: (Aim 1) pathomic MRI biomarkers empha- size the visual differences of aggressive vs. indolent cancers on MRI; (Aim 2) rad-pathomic deep learn- ing models can reliably and automatically distinguish aggressive from indolent prostate cancers on MRI, and (Aim 3) radiologists assisted by deep learning models have increased detection accuracy and inter-reader agreement than unassisted radiologists. Impact: Our proposed rad-pathomic deep learning models have the potential to improve prostate cancer care in three ways: 1) detecting and targeting aggressive cancers that are currently missed in ~50,000 men/year; 2) eliminating up to 500,000 unnecessary biopsies/year in men with no cancer or indolent cancers; and 3) reduc- ing the number of biopsy samples needed to detect aggressive cancers (1-2 vs. 12-18 currently).

项目概要/摘要 前列腺癌是美国男性第二大致命癌症。 MRI 越来越多地用于引导前列腺 活检，每年有可能使 500,000 名男性免受侵入性活检的副作用。然而，微妙 侵袭性、惰性（非致命）癌症与良性组织的 MRI 外观差异产生了三种差异 问题：漏诊癌症、假阳性率高、读者间一致性较差 对话学家。选择性识别侵袭性和惰性癌症对于减少癌症至关重要 死亡，同时最大限度地减少不必要的活检带来的副作用。 我们建议在放射病理性深部开发和使用基于病理学（病理性）MRI 生物标志物 学习帮助放射科医生检测和定位侵袭性癌症与惰性癌症的方法 前列腺 MRI。此外，我们提出的方法将是第一个定位侵袭性和惰性癌症的方法 当它们共存时（76% 的指标病变）。我们在我们独特的数据集中进行了四项初步研究 匹配放射学和病理学图像。首先，我们发现攻击性与惰性的标签高度一致 自动化方法和两名病理学家之间的癌症。其次，我们开发了病理性 MRI 双 来自 MRI 特征的标记，与病理图像的特征相关。第三，我们使用了 辐射致病性深度学习模型中的生物标志物用于检测癌症（AUC：0.86）和侵袭性癌症（AUC： 0.85）在核磁共振成像上。第四，我们展示了将放射科医生和放射病学深度学习模型相结合 帮助发现了放射科医生漏掉的 14% 的侵袭性癌症。 三项创新将改善前列腺 MRI 上侵袭性癌症与惰性癌症的定位。 首先，我们将开发 3D RAPSODI，一种用于 3D 重建 MRI 和病理学的新型 3D 配准方法 图像，以消除切片到切片对应的需要，并将癌症标签从病理学映射到 核磁共振成像。其次，我们将利用相关学习方法来识别病理性 MRI 生物标志物。第三，我们 将使用深度学习模型来协助放射科医生在 MRI 上定位侵袭性癌症。 我们的多学科团队具有独特的优势，可以测试：（目标 1）病理性 MRI 生物标志物是否强调 确定 MRI 上侵袭性癌症与惰性癌症的视觉差异大小； （目标2）rad-pathomic深度学习- 荷兰国际集团的模型可以可靠地、自动地区分侵袭性前列腺癌和惰性前列腺癌 MRI 和（目标 3）放射科医生在深度学习模型的协助下提高了检测准确性和 与无人协助的放射科医生相比，读者间的一致性更高。 影响：我们提出的放射病深度学习模型有可能改善前列腺癌护理 通过三种方式：1）检测和靶向目前每年约 50,000 名男性被忽视的侵袭性癌症； 2） 每年对没有癌症或惰性癌症的男性消除多达 500,000 次不必要的活检；和3）减少 检测侵袭性癌症所需的活检样本数量（目前为 1-2 与 12-18）。