Diffuse Optical B-scan Imaging (DOBI) for Breast Cancer Monitoring

摘要： 组织代谢活动的变化与包括癌症在内的许多疾病有关，但 目前还没有一种非侵入性地、廉价地和高速地对这些变化进行成像的方法。扩散光学 光谱学（DOS）是一种新兴的成像方式，其能够测量血红蛋白浓度， 使用近红外光计算局部血氧饱和度作为代谢活动的代表。然而，在这方面， 当前的DOS探针依赖于手动逐点扫描来建立2D图像。扩散光学层析成像 (DOT)是DOS的深度解析变体，但生成单个图像和设备可能需要数小时 它们本身庞大且复杂。迫切需要一种低成本、非侵入性、深度分辨、高- 通过实时反馈加速代谢成像模式， 与代谢变化相关的疾病，如癌症。如果没有这样的模式，患者将受到 更具侵入性的手术和不必要的治疗，以及令人衰弱的副作用。这个项目的目标是 开发一种临床仪器，产生组织代谢的"超声波"图像，用于评估 局部改变代谢的疾病进展，如癌症和中风。使用最新开发的数字 DOS平台，能够采集速率大于100赫兹，我们将设计和建立一个探头， 单个光源和检测器光纤均以内摆线图案扫描。内摆线是一种形状 当一个小圆绕着一个较大的圆的内圆周滚动时，由一个小圆上的一个点所描绘的。当两 内摆线平行跟踪，点之间的距离在很大范围内变化，而中心 连接两点的直线保持不变。这些独特的功能使DOS测量能够 每个源/检测器间隔对应于在源/检测器间隔下方的不同深度。 组织表面。这种扫描模式将使组织代谢活动的深度分辨测量成为可能。 目前，DOS数据分析速度很慢，需要几秒钟才能分析一个测量结果。增加 数据吞吐率，我们将训练人工神经网络来预测吸收和散射 这将使实时分析的组织的属性。最后，我们将测试新探测器的监测能力， 乳腺癌的血流动力学变化。我们将使用探头对接受新辅助治疗的患者进行成像， 乳腺癌的化疗，并测试探头定位和测量血流动力学的能力， 肿瘤预计该提案将产生以下预期成果：第一，快速，低成本 DOS扫描仪原型将显示能够提供氧合血红蛋白和脱氧血红蛋白的实时图像 与组织代谢有关的浓度。其次，该设备将被证明能够定位乳房 不同患者人群中的肿瘤。第三，这种扫描方法将使乳房的深度分辨图像成为可能。 为今后研究化疗反应铺平了道路。

Abstract: Changes in a tissue’s metabolic activity are associated with many diseases including cancer, yet there is currently no method to image these changes noninvasively, inexpensively, and at high speed. Diffuse Optical Spectroscopy (DOS) is an emerging imaging modality that is able to measure hemoglobin concentrations and calculate local blood oxygen saturation as a proxy for metabolic activity using near-infrared light. However, current DOS probes rely on manual point-by-point scanning to build up 2D images. Diffuse Optical Tomography (DOT) is a depth resolved variant of DOS, but it can take hours to generate a single image and the devices themselves are bulky and complex. There is a critical need for a low-cost, noninvasive, depth resolved, high- speed metabolic imaging modality with real-time feedback to provide frequent monitoring and assessment of diseases associated with metabolic changes such as cancer. Without such a modality, patients are subjected to more invasive procedures and unnecessary treatments with debilitating side-effects. The goal of this project is to develop a clinical instrument that produces “ultrasound-like” images of tissue metabolism for assessing the progression of diseases that locally alter metabolism such as cancer and stroke. Using recently developed digital DOS platform that is capable of acquisition rates greater than 100 Hz, we will design and build a probe in which a single source and detector fiber are each scanned in a hypotrochoidal pattern. A hypotrochoid is a shape traced by a point on a small circle as it rolls around the inner circumference of a larger circle. When two hypocycloids are traced in parallel, the distance between the points varies over a wide range, while the center of the line connecting the two points remains stationary. These unique features enable DOS measurements to be interpreted as an axial line where each source/detector separation corresponds to a different depth below the tissue surface. This scanning pattern will enable depth-resolved measurements of tissue metabolic activity. Currently, DOS data analysis is slow, requiring several seconds to analyze a single measurement. To increase the data throughput rate, we will train an artificial neural network to predict the absorption and scattering properties of the tissue which will enable real-time analysis. Finally, we will test the new probe’s ability to monitor hemodynamic changes in breast cancer. We will use the probe to image patients undergoing neoadjuvant chemotherapy for breast cancer, and test the ability of the probe to locate and measure the hemodynamics of the tumor. It is anticipated this proposal will yield the following expected outcomes: First, a fast, low-cost prototype DOS scanner will be shown capable of providing real-time images of oxy- and deoxy-hemoglobin concentrations which are linked to tissue metabolism. Second, this device will prove capable of locating breast tumors in a diverse patient population. Third, this scanning method will enable depth resolved images of breast tumors paving the way for future work investigating chemotherapy response.