Detection and Grading of Fatty and Fibrotic Liver Using Quantitative Ultrasound

 描述（由申请人提供）：拟议研究的目标是量化定量超声（QUS）询问技术提供的对比度，以将肝脏分类为正常、脂肪、纤维化或脂肪纤维化的组合，并对肝脏中的脂肪或纤维化程度进行分级。脂肪肝是慢性肝病最常见的原因， 越来越多地导致更严重的肝脏病症例如，肝癌、肝硬化或完全肝衰竭。据估计，非酒精性脂肪肝（NAFLD）单独影响 超过30%的美国人和越来越多的肥胖问题在美国，NAFLD将成为一个更加严重的医学问题。目前，脂肪变性（脂肪肝）的准确检测和分类代表了重大的医学挑战。此外，区分脂肪肝与纤维化肝的能力和量化肝脏中纤维化的能力代表了甚至更重要的医学挑战，因为纤维化肝可能更能指示严重的肝脏病症。肝活检仍然是诊断的金标准 弥漫性肝病然而，肝活检及其作为筛查工具的应用存在明显的局限性。实验室和常规放射学检查的使用也被探索用于弥漫性肝病的检测和量化。迄今为止，这些技术也是有限的，因为这些测试不够敏感或特异，无法区分脂肪肝疾病的等级、纤维化的等级，和/或无法区分脂肪肝和纤维化肝。因此，发展能够检测、分类和分级正常肝脏、脂肪肝和纤维化肝的无创技术对医学具有重要意义。 我们在以前的研究中已经证明了QUS根据微观结构特征对疾病进行分类的能力。此外，我们的初步数据表明，QUS可以区分脂肪肝和正常肝脏，并对脂肪变性进行分级。基于这一证据，我们相信QUS将提供能够填补弥漫性肝病非侵入性检测和定量的重要和缺失的作用的能力。为实现这一目标，提出了两个具体目标。第一个具体目标是量化QUS提供的对比度，以在兔模型中对脂肪肝和纤维化肝进行非侵入性分类和分级。通过将先进的QUS询问技术应用于弥漫性肝病兔模型中产生的正常、脂肪和纤维化肝脏，将成功完成这一目标。具体而言，通过5 × 5析因研究设计，我们将探索QUS单独检测和分级正常肝脏、脂肪肝和纤维化肝脏的能力，以及脂肪和纤维化对肝脏疾病的联合影响。第二个具体目标是对具体目标1中提供的QUS参数进行统计分析，用于弥漫性肝病的分类和分级。这一具体目标将通过以下方式实现： 从在第一特定目标中进行的实验中获得的QUS参数在正常、脂肪和纤维化肝脏中聚集以使用线性判别分析创建多参数分类器。将使用多项式约束方法来量化组合饮食和纤维化损伤对肝脏状态的进展。

 DESCRIPTION (provided by applicant): The goal of the proposed research is to quantify the contrast provided by quantitative ultrasound (QUS) interrogation techniques for classifying liver as normal, fatty, fibrotic or a combination of fatty fibrotic and to grade the degree of fat or fibrosis in the liver. Fatty liver disease is the most common cause of chronic liver disease and is increasingly leading to more severe liver conditions e.g., hepatocarcinoma, cirrhosis, or complete liver failure. It is estimated that nonalcoholic fatty liver disease (NAFLD) alone affects over 30% of Americans and with increasing problems with obesity in the U.S., NAFLD is poised to become an even more serious medical concern. At present, accurate detection and classification of steatosis (fatty liver) represents a significant medical challenge. Furthermore, the ability to differentiate fatty liver from fibrotic liver and the ability to quantify fibrosis i the liver represents an even more significant medical challenge because the fibrotic liver may be more indicative of severe liver conditions. Liver biopsy continues to be the gold standard for diagnosis of diffuse liver disease. However, there are significant limitations associated with liver biopsy and its use as a screening tool. The use of laboratory and conventional radiological tests has also been explored for the detection and quantification of diffuse liver disease. To date these techniques are also limited because these tests are not sensitive or specific enough to distinguish the grade of fatty liver disease, the grade of fibrosis, and/or are unable to distinguih between fatty and fibrotic liver. Therefore, it is of great importance to medicine to develop noninvasive techniques capable of detecting, classifying and grading normal, fatty and fibrotic liver. We have demonstrated in previous studies the ability of QUS to classify disease based on microstructural features. In addition, our preliminary data suggest that QUS can differentiate fatty liver from normal liver and grade steatosis. Based on this evidence, we believe QUS will provide capabilities that can fill an important and missing role in noninvasive detection and quantification of diffuse liver disease. To accomplish this objective, two specific aims are proposed. The first specific aim is to quantify contrast provided by QUS to noninvasively classify and grade both fatty liver and fibrotic liver in a rabbit model. This aim will be successfully completed by applying advanced QUS interrogation techniques in vivo to normal, fatty, and fibrotic livers produced in rabbit models of diffuse liver disease. Specifically, through a 5 by 5 factorial study design we will explore the ability of QUS to detect and grade normal, fatty, and fibrotic liver alone and the combined effects of fat and fibrosis on liver disease. The second specific aim is to provide statistical analysis of the QUS parameters provided in specific aim 1 for classification and grading of diffuse liver disease. This specific aim will be met by combining the QUS parameters acquired from the experiments conducted in the first specific aim in aggregate in normal, fatty and fibrotic liver to create a multiparameter classifier using linear discriminant analysis. A polynomial constraints method will be used to quantify the progression of combined diet and fibrotic injury on liver state.