Estimation and Correction of Ultrasound Beam Aberration Caused by Breast

描述（由申请人提供）：本研究的目的是开发超声传播算法和组织模型，模拟典型乳腺标本中的超声散射行为，用于研究乳腺成像技术，然后利用这些模型和算法来改进和评估自适应聚焦方法，以纠正由于乳房不均匀性传播引起的超声束畸变。具体目标是：1)在整个乳房的体积内获取高分辨率的磁共振成像数据，并将数据分割成具有随机过程特征的声学性质的组织类型；2)通过模拟乳房的体积计算三维声学传播，以确定乳房内不均匀性产生的像差；3)使用相同的样品进行脉冲回波测量，以验证乳房的建模和像差的计算；4)通过使用补偿像差的自适应聚焦模拟高分辨率b扫描图像。分割后的数据将用于发展实际的数值模型来计算传播。通过乳房模型计算脉冲传播将使用三维k空间和快速多重方法。通过模型的传播将用于模拟可重复且易于更改的测量，从而允许有效地改进像差校正方法。像差将使用两种新的算法来确定。在一种算法中，利用从等平面区域的一组焦点获得的脉冲回波信号的交叉谱估计像差。在另一种算法中，利用虚拟光源产生的宽光束照明回波的互相关估计像差。我们将探讨控制这些算法的参数的变化，以优化算法的性能。通过像差路径接收的模拟点反射回波将用于计算真实像差，并与使用两种算法得到的像差进行比较。将通过计算和水听器测量来描述每种情况下获得的真正焦点。焦点特征以及图像分辨率将根据传播路径中的乳腺形态进行评估。将根据计算和测量结果确定到达时间波动、波形形状变化和失真统计。此外，还将确定用一组参数可以满意地补偿像差的区域的大小。焦点特性和图像分辨率将仔细评估，并严格比较现有的几何和自适应聚焦技术。定量结论的影响，像差在超声成像的乳房和性能的两种像差校正算法将发展。作为这项研究的结果，使用像差校正的自适应聚焦技术将产生更高的分辨率，这将显著提高超声b扫描成像的能力，以区分正常和患病的乳房组织，并在目前超声无法实现的情况下确定乳房疾病的严重程度。

DESCRIPTION (provided by applicant): The objective of this research is to develop ultrasound propagation algorithms and tissue models that mimic ultrasound scattering behavior in typical breast specimens for use in studying breast imaging techniques, and then to employ these models and algorithms to refine and evaluate adaptive focusing methods that correct for ultrasound beam aberration caused by propagation through breast inhomogeneities. The specific aims are to: 1) acquire high-resolution magnetic resonance imaging data throughout the volume of the breast and segment the data into tissue types with acoustic properties characterized by random processes, 2) calculate acoustic propagation in three dimensions through the modeled volume of the breast to determine the aberration produced by inhomogeneities in the breast, 3) perform pulse-echo measurements of aberration using the same specimens to validate the modeling of the breast and the calculation of aberration, and 4) simulate high-resolution b-scan images by using adaptive focusing that compensates for aberration. The segmented data will be used to develop realistic numerical models for calculations of propagation. Calculations of pulse propagation through the breast models will use three-dimensional k-space and fast multiple methods. Propagation through the models will be used to simulate measurements that are repeatable and easy to alter, allowing for efficient refinement of aberration-correction methods. Aberration will be determined using two new algorithms. In one algorithm, the aberration is estimated using cross spectra of pulse-echo signals obtained from a set of focuses in an isoplanatic region. In the other algorithm, aberration is estimated using cross correlation of echoes obtained from broad-beam illuminations produced by virtual sources. Variations of the parameters that govern these algorithms will be explored to optimize algorithm performance. Simulated point-reflector echoes received through the aberration path will be used to compute the true aberration for comparison with the aberration found using the two algorithms. The true focus achieved in each of the cases will be described by calculations and also by hydrophone measurements. Focus characteristics as well as image resolution will be evaluated with respect to breast morphology in the propagation paths. Arrival time fluctuations, waveform shape changes, and statistics of distortion will be determined for both calculated and measured results. Also, the size of the region over which aberration can be satisfactorily compensated with a single set of parameters will be determined. Focus characteristics and image resolution will be carefully evaluated and critically compared to available geometric and adaptive focusing techniques. Quantitative conclusions about the effects of aberration in ultrasound imaging of the breast and the performance of the two aberration correction algorithms will be developed. As a result of this research, adaptive focusing techniques using aberration correction will yield improved resolution that will significantly increase the capability of ultrasound b-scan imaging to distinguish between normal and diseased breast tissue and to determine the severity of breast disease in circumstances not now possible with ultrasound. PUBLIC HEALTH RELEVANCE: The objective of this project is to form significantly improved ultrasonic images throughout the volume of the breast by using adaptive focusing that compensates for aberration. This objective will be achieved by the development of realistic acoustic models of the breast from high-resolution magnetic resonance imaging data, use of the models to estimate aberration, validation of the models by measurements, and formation of b-scan images by using aberration correction. Success in the research will significantly increase the capability of ultrasound b-scans to distinguish between normal and diseased breast tissue and to determine the severity of disease in circumstances not now possible because b-scan resolution in breast is limited by aberration.