Image Quality Improvement and Performance Assessment of Dedicated Breast CT

专用乳腺CT的图像质量改进和性能评估

基本信息

批准号：
9899937
负责人：
Ioannis Sechopoulos
金额：
$ 19.66万
依托单位：
STICHTING RADBOUD UNIVERSITAIR MEDISCH CENTRUM I.O.
依托单位国家：
美国
项目类别：
财政年份：
2015
资助国家：
美国
起止时间：
2015-04-01 至 2023-03-31
项目状态：
已结题

来源：
https://reporter.nih.gov/project-details/9899937
关键词：
3-Dimensional Algorithms American Cancer Society Anatomy Biopsy Breast Breast Cancer Detection Breast Cancer Early Detection Breast Cancer Patient Breast Microcalcification Cancer Diagnostics Characteristics Clinical Clinical Trials Contrast Media Development Diagnosis Diagnostic Diagnostic Procedure Disease Early Diagnosis Ensure Healthcare Image Image Analysis Imaging technology Information Distribution Injections Kinetics Lesion Malignant Neoplasms Mammography Methods Modality Morphologic artifacts Nature Neoadjuvant Therapy Patient imaging Patients Performance Physicians Probability Process Radiopharmaceuticals Resolution Roentgen Rays Shapes Signal Transduction Survival Rate Testing Three-Dimensional Image Time Tissues Ultrasonography Visualization Woman Work X-Ray Computed Tomography base breast cancer diagnosis breast density breast exam breast imaging cancer imaging chemotherapy clinical application clinical diagnostics clinical efficacy clinical imaging contrast enhanced density detector diagnostic accuracy digital imaging image processing image reconstruction imaging capabilities imaging modality imaging study improved malignant breast neoplasm novel novel imaging technology prospective public health relevance quantitative imaging radiologist reconstruction response tomography tomosynthesis tumor two-dimensional

项目摘要

DESCRIPTION (provided by applicant): The diagnostic stage of early detection of breast cancer is currently far from perfect. With the current clinical imaging technology used for diagnostic work-up of suspicious lesions detected at breast cancer screening, approximately one out of six breast cancers are missed. This is in addition to the one out of six breast cancers already missed at the breast cancer screening stage. In addition, the rate of false positive diagnostic work-ups results in more than two out of three biopsies yielding a negative result. Clearly there is room for improvement in this very important clinical diagnostic procedure. Of the many novel imaging technologies being developed for breast cancer imaging, dedicated breast computed tomography (BCT) is one of very few that results in a true tomographic image of the breast with high contrast resolution and does not require the injection of a contrast agent or radiopharmaceutical. This makes it ideal for use as the frontline imaging technology for working up suspicious lesions detected during breast cancer screening or during clinical breast examination. In this project we propose to perform a prospective clinical trial to compare the accuracy of BCT to the current standard imaging technologies for diagnosis of breast cancer in patients with a suspicious lesion identified during breast cancer screening. To maximize the image quality of BCT we will apply to the BCT novel algorithms we have developed that result in more accurate, higher quality images with true quantitative characteristics. These algorithms involve the correction of the acquired BCT projections due to the presence of x-ray scatter, and a novel reconstruction algorithm that correctly represents the image acquisition process as one involving a spectrum of x-ray energy, rather than mono-energetic x-rays. Successful completion of this project will help introduce BCT to the clinical realm by characterizing its true potential or impact in the breast cancer diagnosis stage, where we expect it will result in fewer missed breast cancers and negative biopsies.

描述（由适用提供）：乳腺癌早期发现的诊断阶段目前远非完美。由于目前使用的临床成像技术用于诊断乳腺癌筛查中可疑水平的诊断检查，因此遗失了六分之一的乳腺癌。这是在乳腺癌筛查阶段已经错过的六个乳腺癌中的一种。此外，假阳性诊断检查的速率导致三个活检中有超过两次产生负面结果。显然，这种非常重要的临床诊断程序有改善的余地。在为乳腺癌成像开发的许多新型成像技术中，专用的乳房计算机断层扫描（BCT）是很少的乳腺层析成像之一，它导致具有高对比度分辨率的乳房的真正层析成像图像，并且不需要注射对比剂或放射性药物。这使其非常适合用作乳腺癌筛查或临床乳房检查期间检测到的可疑病变的一线成像技术。在该项目中，我们建议进行一项前瞻性临床试验，以将BCT的准确性与当前标准成像技术进行乳腺癌诊断，以抑制乳腺癌筛查期间鉴定出可疑病变的患者。为了最大化BCT的图像质量，我们将应用于BCT新颖的算法，我们开发了具有真正定量特征的更准确，更高质量的图像。这些算法涉及由于存在X射线散射而校正所获得的BCT项目，以及一种新颖的重建算法，该算法正确地代表了图像采集过程是涉及一系列X射线能量的X射线能量，而不是单一能量X射线。该项目的成功完成将通过表征其在乳腺癌诊断阶段的真正潜力或影响来帮助将BCT引入临床领域，我们预计这将导致较少的乳腺癌和阴性活检。

项目成果

期刊论文数量（12）

专著数量（0）

科研奖励数量（0）

会议论文数量（0）

专利数量（0）

Development of 3D patient-based super-resolution digital breast phantoms using machine learning.

DOI：
10.1088/1361-6560/aae78d
发表时间：
2018-11-12
期刊：
Physics in medicine and biology
影响因子：
3.5
作者：
Caballo M;Fedon C;Brombal L;Mann R;Longo R;Sechopoulos I
通讯作者：
Sechopoulos I

Deep learning-based segmentation of breast masses in dedicated breast CT imaging: Radiomic feature stability between radiologists and artificial intelligence.

DOI：
10.1016/j.compbiomed.2020.103629
发表时间：
2020-03
期刊：
Computers in biology and medicine
影响因子：
7.7
作者：
通讯作者：

The compressed breast during mammography and breast tomosynthesis: in vivo shape characterization and modeling.

DOI：
10.1088/1361-6560/aa7cd0
发表时间：
2017-08-07
期刊：
Physics in medicine and biology
影响因子：
3.5
作者：
Rodríguez-Ruiz A;Agasthya GA;Sechopoulos I
通讯作者：
Sechopoulos I

Deep learning reconstruction of digital breast tomosynthesis images for accurate breast density and patient-specific radiation dose estimation.

数字乳腺断层合成图像的深度学习重建，用于准确的乳腺密度和患者特定的辐射剂量估计。

DOI：
10.1016/j.media.2021.102061
发表时间：
2021
期刊：
Medical image analysis
影响因子：
10.9
作者：
Teuwen,Jonas;Moriakov,Nikita;Fedon,Christian;Caballo,Marco;Reiser,Ingrid;Bakic,Pedrag;García,Eloy;Diaz,Oliver;Michielsen,Koen;Sechopoulos,Ioannis
通讯作者：
Sechopoulos,Ioannis

An unsupervised automatic segmentation algorithm for breast tissue classification of dedicated breast computed tomography images.