MR Fingerprinting and Computerized Decision Support for Prostate Cancer

前列腺癌的 MR 指纹识别和计算机化决策支持

• 批准号: 10219975
• 负责人: Vikas Gulani
• 金额: $ 58.17万
• 依托单位: UNIVERSITY OF MICHIGAN AT ANN ARBOR
• 依托单位国家: 美国
• 财政年份: 2017
• 资助国家: 美国
• 起止时间: 2017-03-02 至 2024-02-29
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10219975
• 关键词: Algorithmic Analysis; Algorithms; Benign; Biopsy; Cancer Patient; Clinical; Complement; Data; Detection; Development; Diagnosis; Diffusion; Discrimination; Disease; Evaluation; Exclusion; Failure; Fingerprint; Goals; Heterogeneity; Image; Image Analysis; Imaging Techniques; Imaging technology; Indolent; Learning; Lesion; Literature; Magnetic Resonance; Magnetic Resonance Imaging; Malignant Neoplasms; Malignant neoplasm of prostate; Maps; Measurement; Monitor; Morphology; Neoplasms; Operative Surgical Procedures; Organ; Outcome; Palpable; Patients; Pattern; Performance; Periodicity; Peripheral; Probability; Property; Prostate; Prostate-Specific Antigen; Protocols documentation; Publishing; Risk; Scanning; Site; Slice; Stress; Technology; Testing; Texture; Tissues; Treatment Protocols; Validation; Variant; Visual; Work; aggressive therapy; base; blind; cancer heterogeneity; cohort; computerized; cost; disorder risk; improved; innovation; men; patient subsets; prospective; prostate biopsy; prostatitis; radiomics; support tools; tool; tumor

PROJECT SUMMARY The prostate remains the only organ in which blind untargeted biopsies are conducted without a pre-identified suspected focus of neoplasm. Men with palpable abnormalities or elevated prostate specific antigen must endure trans-rectal biopsy with related costs, discomfort, stress, and complications, because it is not impossible to objectively and reliably identify who does not require a biopsy. Well over 500,000 men in the US with no evidence of prostate cancer still undergo prostate biopsies, solely on account of PSA, a grade D test according to the USPTF. Consequently there is clearly an unmet need to develop both better imaging techniques and image analysis algorithms that can enable improved non-invasive characterization of prostate cancer and distinguish low grade indolent cancers from the more aggressive intermediate to high grade variants. This would help channel and monitor appropriate patients in less aggressive treatment protocols such as active surveillance. Currently MRI is excellent for detecting high grade prostate cancer (PCa), but is less accurate for low and intermediate grade disease. Definitive exclusion of disease, and thus the need for biopsy in a subset of patients, is not possible. Also, patients who opt for active surveillance cannot be followed by imaging alone and require repeated periodic biopsy. Magnetic resonance fingerprinting (MRF) is a framework pioneered by our team for simultaneously quantifying multiple tissue properties with MRI, and has been used to quantify T1 and T2 more efficiently, accurately, and precisely than previously possible. Extensive preliminary data show the utility of this technology in combination with apparent diffusion coefficient (ADC) mapping, to separate normal peripheral zone from potential cancer. In parallel our team has been developing and validating computerized decision support (CDS) tools which can diagnose, grade, and characterize PCa both in the peripheral and transitional zones on MRI. We propose to develop an MRF exam for prostate cancer that allows simultaneous mapping of T1, T2, and ADC for efficient and quantitative separation of PCa from normal prostate and to separate low risk and more aggressive disease. We will also develop integrated CDS tools to identify additional image derived features (radiomics) from the MRF derived maps to complement MRF measurements for more accurate detection and grading of PCa both in the peripheral and transitional zones. The accuracy of this combined MRF+CDS exam will be prospectively validated in a cohort of 250 men scanned prior to biopsy.

项目摘要 前列腺仍然是唯一的器官，其中盲非靶向活检进行没有预先确定的 疑似肿瘤病灶。可触及异常或前列腺特异性抗原升高的男性必须 忍受经直肠活检的相关费用，不适，压力和并发症，因为它不是 不可能客观可靠地确定谁不需要活检。美国有超过50万人 没有前列腺癌的证据仍然接受前列腺活检，仅仅是因为PSA，一个D级测试 根据USPTF。因此，显然存在未满足的需求，即开发更好的成像技术， 可以改进前列腺非侵入性表征的技术和图像分析算法 癌症和区分低级别惰性癌症从更具侵略性的中级到高级 变体。这将有助于引导和监测适当的患者在不太积极的治疗方案， 主动监视 目前，MRI对于检测高级别前列腺癌（PCa）是极好的，但是对于低级别和高级别前列腺癌（PCa）是不太准确的。 中度疾病。排除疾病，因此需要在一个子集的活检， 病人是不可能的。此外，选择主动监测的患者不能仅通过成像进行随访， 需要反复定期活检。磁共振指纹识别（MRF）是由我们的团队开创的一个框架。 团队同时量化多种组织特性与MRI，并已被用于量化T1和 T2比以前更有效、更准确、更精确。大量的初步数据显示， 该技术与表观扩散系数（ADC）标测相结合的实用性， 周围区域的潜在癌症。同时，我们的团队一直在开发和验证计算机化的 决策支持（CDS）工具，可以诊断、分级和表征外周和外周的PCa 移行区的影像我们建议开发一种前列腺癌的MRF检查， 同时绘制T1、T2和ADC图，用于有效和定量分离PCa与正常人 前列腺和分离低风险和更具侵略性的疾病。我们还将开发综合CDS 从MRF衍生图中识别其他图像衍生特征（放射组学）的工具， 补充MRF测量，以便更准确地检测和分级外周和外周中的PCa 和过渡区。这种MRF+CDS联合检查的准确性将在 活检前扫描的250名男性队列。

项目成果

Computer-Aided Laser Dissection: A Microdissection Workflow Leveraging Image Analysis Tools.

• DOI: 10.4103/jpi.jpi_60_18
• 发表时间: 2018-01-01
• 期刊: Journal of pathology informatics
• 作者: Hipp, Jason D;Johann, Donald J;Tangrea, Michael A
• 通讯作者: Tangrea, Michael A

Accurate and reproducible invasive breast cancer detection in whole-slide images: A Deep Learning approach for quantifying tumor extent.

• DOI: 10.1038/srep46450
• 发表时间: 2017-04-18
• 期刊: Scientific reports
• 影响因子: 4.6
• 作者: Cruz-Roa A;Gilmore H;Basavanhally A;Feldman M;Ganesan S;Shih NNC;Tomaszewski J;González FA;Madabhushi A
• 通讯作者: Madabhushi A

CT radiomic signature predicts survival and chemotherapy benefit in stage I and II HPV-associated oropharyngeal carcinoma.

• DOI: 10.1038/s41698-023-00404-w
• 发表时间: 2023-06-02
• 期刊: NPJ precision oncology
• 影响因子: 7.9

Intratumoral and peritumoral radiomics for the pretreatment prediction of pathological complete response to neoadjuvant chemotherapy based on breast DCE-MRI.

• DOI: 10.1186/s13058-017-0846-1
• 发表时间: 2017-05-18
• 期刊: Breast cancer research : BCR
• 作者: Braman NM;Etesami M;Prasanna P;Dubchuk C;Gilmore H;Tiwari P;Plecha D;Madabhushi A
• 通讯作者: Madabhushi A

High-throughput adaptive sampling for whole-slide histopathology image analysis (HASHI) via convolutional neural networks: Application to invasive breast cancer detection.

• DOI: 10.1371/journal.pone.0196828
• 发表时间: 2018
• 期刊: PloS one
• 影响因子: 3.7
• 作者: Cruz-Roa A;Gilmore H;Basavanhally A;Feldman M;Ganesan S;Shih N;Tomaszewski J;Madabhushi A;González F
• 通讯作者: González F