Software tool for routine, rapid, patient-specific CT organ dose estimation

用于常规、快速、患者特定 CT 器官剂量估算的软件工具

• 批准号: 9922675
• 负责人: Taly Gilat Schmidt
• 金额: $ 43.28万
• 依托单位: MARQUETTE UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2017
• 资助国家: 美国
• 起止时间: 2017-07-01 至 2023-03-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/9922675
• 关键词: Abdomen; Accounting; Acute; Address; Adoption; Algorithms; Anatomy; Atlases; Case Study; Chest; Childhood; Clinical; Collaborations; Computed Tomography Scanners; Computerized Medical Record; Data; Data Set; Databases; Dose; Engineering; Equation; Future; Goals; Gold; Image; Incidence; Individual; Industrialization; Informatics; Injury; International; Ionizing radiation; Malignant Neoplasms; Manuals; Maps; Medical; Medicine; Methods; Modeling; Monitor; National Institute of Biomedical Imaging and Bioengineering; Organ; Overdose; Patients; Pelvis; Physics; Population; Procedures; Protocols documentation; Public Health; Radiation; Radiation Dose Unit; Radiation Oncology; Radiation therapy; Radiology Specialty; Regulation; Reporting; Research; Resources; Running; Scanning; Software Tools; System; Time; Tube; Variant; Work; X-Ray Computed Tomography; automated segmentation; base; cancer risk; clinical practice; commercialization; epidemiology study; experimental study; indexing; innovation; novel; pediatric patients; phantom model; prevent; prototype; public health relevance; segmentation algorithm; simulation; software development; symposium; tool

PROJECT SUMMARY The approximately 76 million Computed Tomography (CT) scans performed in the U.S. each year are responsible for half the ionizing radiation delivered by medical procedures. Concern about stochastic cancer risks and recent overdosing incidents has led to increased radiation dose monitoring and mandated dose reporting in several states. The problem addressed by this proposal is that current dose reporting metrics quantify the dose delivered to a plastic cylinder or dose to a phantom model, not the dose to the organs of a specific patient. Numerous national and international reports have identified individual organ dose as the most relevant metric that should be reported. Existing automated tools do not model the patient's anatomy and have >40% organ dose error for some reported cases. This project will develop an automated software tool to provide the new capability of accurate, rapid, and personalized reporting of the radiation dose delivered to a patient's organs as part of every CT scan. This project will leverage expertise from the radiology and radiation oncology fields to develop innovative algorithms that will provide the new capability of personalized CT organ dose estimates that account for scanner and anatomical complexities. To achieve the project aims, a rapid Boltzmann Transport Equation solver will be optimized for CT imaging, expanded to model scanner complexities, and validated against gold-standard simulations and phantom experiments. Automated atlas-based segmentation algorithms will be developed, validated, and combined with novel methods to robustly estimate organ dose. The complete dose estimation tool will be validated in a study of 500 pediatric CT datasets, which will provide valuable information about the magnitude and variation of pediatric CT dose in clinical practice. The resulting organ-dose database will be made publically available as a resource for clinical and technical research. The expected impact of the proposed software tool is: (1) Patient-specific organ doses and dose maps incorporated into electronic medical records for personalized dose reports. (2) Personalized dose minimization when combined with dynamic filters and adaptive tube current modulation. (3) Databases for protocol optimization and epidemiological studies of organ dose and cancer incidence based on accurate dose estimates that quantify the variation in organ dose across the population.

项目概要 美国每年进行约 7600 万次计算机断层扫描 (CT) 造成医疗程序产生的电离辐射的一半。对随机癌症的担忧 风险和最近的用药过量事件导致加强了辐射剂量监测和强制剂量 在几个州进行报道。该提案解决的问题是当前的剂量报告指标量化 传递到塑料圆筒或模型模型的剂量，而不是特定器官的剂量 病人。许多国家和国际报告已确定个体器官剂量是最相关的 应报告的指标。现有的自动化工具无法模拟患者的解剖结构，并且具有 >40% 一些报告病例的器官剂量错误。该项目将开发一个自动化软件工具来提供 准确、快速和个性化地报告传递到患者器官的辐射剂量的新功能 作为每次 CT 扫描的一部分。 该项目将利用放射学和放射肿瘤学领域的专业知识来开发创新 算法将提供个性化 CT 器官剂量估计的新功能（考虑到扫描仪） 和解剖学的复杂性。为了实现该项目目标，将使用快速玻尔兹曼传输方程求解器 针对 CT 成像进行了优化，扩展到模型扫描仪复杂性，并根据黄金标准进行了验证 模拟和幻象实验。将开发基于图集的自动分割算法， 经过验证，并与新方法相结合，可以稳健地估计器官剂量。完整的剂量估算工具 将在 500 个儿科 CT 数据集的研究中得到验证，这将提供有关该疾病的宝贵信息 临床实践中儿科 CT 剂量的大小和变化。由此产生的器官剂量数据库将被建立 作为临床和技术研究的资源公开提供。 所提出的软件工具的预期影响是：（1）患者特定器官剂量和剂量图 纳入电子病历以提供个性化剂量报告。 (2)个性化剂量最小化 与动态滤波器和自适应管电流调制结合使用时。 (3) 协议数据库 基于准确的剂量估计的器官剂量和癌症发病率的优化和流行病学研究 量化整个人群器官剂量的变化。

项目成果

Deterministic linear Boltzmann transport equation solver for patient-specific CT dose estimation: Comparison against a Monte Carlo benchmark for realistic scanner configurations and patient models.

• DOI: 10.1002/mp.14494
• 发表时间: 2020-12
• 期刊: Medical physics
• 影响因子: 3.8
• 作者: Principi S;Wang A;Maslowski A;Wareing T;Jordan P;Schmidt TG
• 通讯作者: Schmidt TG

Pediatric chest-abdomen-pelvis and abdomen-pelvis CT images with expert organ contours.

• DOI: 10.1002/mp.15485
• 发表时间: 2022-05
• 期刊: Medical physics
• 影响因子: 3.8

Reduced Chest Computed Tomography Scan Length for Patients Positive for Coronavirus Disease 2019: Dose Reduction and Impact on Diagnostic Utility.

• DOI: 10.1097/rct.0000000000001312
• 发表时间: 2022-07-01
• 期刊: Journal of computer assisted tomography
• 影响因子: 1.3

Technical note: Evaluation of a V-Net autosegmentation algorithm for pediatric CT scans: Performance, generalizability, and application to patient-specific CT dosimetry.

• DOI: 10.1002/mp.15521
• 发表时间: 2022-04
• 期刊: MEDICAL PHYSICS
• 影响因子: 3.8
• 作者: Adamson, Philip M.;Bhattbhatt, Vrunda;Principi, Sara;Beriwal, Surabhi;Strain, Linda S.;Offe, Michael;Wang, Adam S.;Vo, Nghia-Jack;Schmidt, Taly Gilat;Jordan, Petr
• 通讯作者: Jordan, Petr

AGE-CONDITIONED SYNTHESIS OF PEDIATRIC COMPUTED TOMOGRAPHY WITH AUXILIARY CLASSIFIER GENERATIVE ADVERSARIAL NETWORKS.

• DOI: 10.1109/isbi45749.2020.9098623
• 发表时间: 2020-04
• 期刊: Proceedings. IEEE International Symposium on Biomedical Imaging
• 作者: Kan CNE;Maheenaboobacker N;Ye DH
• 通讯作者: Ye DH