High-resolution, quantitative whole-breast ultrasound: bridging the gap to clinical practice

 描述（由申请人提供）：SoftVue超声断层扫描（UST）系统通过将体积成像能力与成本效益、无电离辐射和操作员独立性相结合，提供了一种对整个乳房进行成像的解决方案。我们的长期目标是开发成像系统，竞争对手的成像性能的MRI，以提供一个低成本的替代MRI和扩大的作用，在乳腺癌筛查。在第一阶段，我们证明了，基于数值模型，波形断层扫描可以实现类似MRI的图像的分辨率，并可以在临床相关的时间尺度上实施。在随后的初步研究中，我们应用波形断层扫描在体内数据。本II期研究的目的是完善波形断层扫描，并将其集成到准备产品化的临床软件包中，以支持我们的商业化计划。我们的中心假设是，集成波形断层扫描将能够产生在体内乳腺图像与MRI的质量在临床相关的时间尺度。将通过对30例患者的研究来检验该假设，这些患者通过标准治疗被确定为具有可疑病变，并且还具有可用于比较的MR图像。预期的结果是对所提出的技术进行临床验证，这反过来又将支持SoftVue在短期内作为乳腺成像的诊断模式，并在长期内作为筛查模式。这项研究的基本原理是，它将推动SoftVue实现我们的目标，消除成像成本和成像性能之间的困境，并支持SoftVue系统的商业化。我们建议的研究将通过以下具体目标进行。目的1：评估所提出的技术对SoftVue系统采集数据的性能。目标二：实施第一阶段研究中新SoftVue刀片服务器的设计，以反映CPU和GPU的最新架构，从而最大限度地提高计算能力。目标3：使用体内数据在新的SoftVue刀片服务器架构上实施和测试所提出的算法，并评估临床应用的计算速度。目标4：将算法与SoftVue产品打包，并支持SoftVue的FDA申请。

 DESCRIPTION (provided by applicant): The SoftVue ultrasound tomography (UST) system provides a solution for imaging the whole breast by combining volumetric imaging capability with cost-efficacy, absence of ionizing radiation and operator independence. Our long term goal is to develop imaging system that rivals the imaging performance of MRI in order to provide a low cost alternative to MRI and an expanded role in breast screening. In Phase I we demonstrated, based on numerical models, that waveform tomography can achieve MRI-like images in terms of resolution and can be implemented on clinically relevant time scales. In subsequent preliminary studies we have applied waveform tomography to in vivo data. The objective of this Phase II study is to refine waveform tomography and integrate it into a clinical software package that would be ready for productization, in support of our commercialization plan. Our central hypothesis is that the integrated waveform tomography will be able to produce in vivo breast images with MRI-like quality on clinically relevant time scales. The hypothesis will be tested via a study of 30 patients who have been identified by standard of care to have suspicious lesions and also have MR images available for comparison. The expected outcome is a clinical validation of the proposed technique, which, in turn, will support SoftVue's role as a diagnostic modality for breast imaging, in the short term, and as a screening modality, in the long term. The rationale for this study is that it will advance SoftVue to achieve our goal of eliminating the dilemma between imaging cost and imaging performance, and support commercialization of the SoftVue system. Our proposed study will be carried out via the following specific aims. Aim 1: Assess the performance of the proposed technique on data acquired with the SoftVue system. Aim 2: Implement design of the new SoftVue blade server from the Phase I study to reflect state-of-the-art architecture for both CPUs and GPUs and thereby maximize the computational capacity. Aim 3: Implement and test the proposed algorithm on the new SoftVue blade server architecture using in vivo data and assess the computational speed for clinical applications. Aim 4: Package the algorithm with the SoftVue product and support SoftVue's FDA applications.