Magnetic Resonance Spectroscopy of Breast Cancer using Model-Based Localization

描述（由申请人提供）：乳腺癌是最常见的女性癌症，仅在美国每年就有超过225，000例新发病例和近40，000例死亡。磁共振成像（MRI）已被确立为表征乳腺癌的最准确的成像方式，现在广泛用于高风险筛查，诊断和评估治疗反应。磁共振波谱（MRS）是一种可以添加到MRI扫描中并提供有关病变的额外代谢信息的技术。研究表明，测量乳腺癌中总胆碱（tCho）的浓度可以提供非常有价值的临床信息。tCho在癌症中的浓度高于良性肿块，因此tCho可用于提高诊断准确性。研究还表明，在接受化疗的患者中，在开始治疗后1天测量的tCho下降与良好的治疗结局相关。目前有两种技术可用于测量乳腺癌中的tCho。最常用的技术称为单体素光谱（SVS）。SVS通常可以产生高质量的数据，从而能够准确测量tCho，但它需要经过专门培训的专家在场并在扫描期间进行调整。在临床实践中，这是非常具有挑战性的，并且一直是使SVS成为实用临床工具的关键障碍。另一种技术称为化学位移成像（CSI）。这具有在扫描时不需要专家的优点，并且它对于研究大脑和前列腺等器官表现得非常好。但在乳房中，来自脂肪组织的大脂质信号的存在会在CSI数据中产生伪影，使得tCho的准确测量不太实用。在这项资助中，我们建议使用一种称为SLIM（成像光谱定位）的旧技术，并以一种新的创新方式使用它来获取与SVS相同或更好质量的数据，但不需要专业操作员。这项创新的关键是将SLIM与现代水脂肪成像方法相结合，可以获得乳房中脂肪和水的高分辨率定量图。然后SLIM可以使用这种高分辨率信息将脂肪信号从水信号中分离出来，而水含有具有临床价值的tCho。在这项资助中，我们将沿着发展SLIM技术以及几种有前途的SLIM变体。我们将通过进行计算机模拟、在复制乳房组成的幻影上进行实验以及在正常女性志愿者中进行实验来完善和测试这些方法。一旦技术得到充分完善，我们将在已经接受MRI扫描的乳腺癌患者中测量tCho，以测试新方法的可行性，并将其与CSI和SVS进行比较。如果这项创新技术开发研究取得成功，我们将创造一种新的非侵入性工具，放射科医生和肿瘤科医生可以用来指导乳腺癌女性的护理和治疗。

DESCRIPTION (provided by applicant): Breast cancer is the most common female cancer, with more than 225,000 new cases and nearly 40,000 deaths per year in the US alone. Magnetic resonance imaging (MRI) has become established as the most accurate imaging modality for characterizing breast cancer, and is now widely used for high-risk screening, diagnosis, and assessing treatment response. Magnetic resonance spectroscopy (MRS) is a technique that can be added to an MRI scan and provide addition metabolic information about a lesion. Studies have shown that measuring the concentration of total choline (tCho) in a breast cancer can provide very valuable clinical information. The tCho concentration is higher in cancers than in benign masses, and therefore tCho can be used to improve diagnostic accuracy. It has also been shown that in patients receiving chemotherapy, a drop in tCho measured as soon as 1 day after starting treatment is associated with good treatment outcomes. There are two techniques currently available to measure tCho in breast cancers. The most commonly used technique is called single-voxel spectroscopy (SVS). SVS generally can produce high-quality data the enables accurate measurement of tCho, but it requires that a specially trained expert is present and making adjustments during the scan. In clinical practice, this is very challenging, and has been a critical barrier to making SVS a practical clinical tool. Another technique is called chemical shift imaging (CSI). This has the advantage of not need an expert at scan time, and it performs very well for studying organs such as the brain and prostate. But in the breast, the presence of large lipid signals from adipose tissue produces artifacts in CSI data, making accurate measurements of tCho less practical. In this grant we propose to use an old technique, called SLIM (spectroscopic localization by imaging), and use it in a new and innovative way to acquire data with the same or better quality as SVS, but without the need for the expert operator. The key to this innovation is combining SLIM with modern water-fat imaging methods, which can acquire high-resolution, quantitative maps of the fat and water in the breast. SLIM can then use this high-resolution information to separate the fat signals from the water signals, and the water contains the tCho that is so clinically valuable. In this grant we will develop the SLIM technique along with several promising variations of SLIM. We will refine and test these methods by performing computer simulations, experiments on phantoms that replicate the breast composition, and in normal female volunteers. Once the techniques are sufficiently refined, we will measure tCho in breast cancer patients who are already receiving MRI scans to test the feasibility of the new methods and compare them CSI and SVS. If successful in this innovative technical development study, we will create a new, non-invasive tool that radiologists and oncologists can use to guide the care and treatment of women with breast cancer.