Space-Time Microwave Imaging for Breast Cancer Detection

描述(由申请人提供)：建议研究的目标是 论证超宽带时空微波成像的可行性 (STMI)方法用于早期乳腺癌检测。选举的结果是 建议的工作将是1)关于信号处理的健壮扫描隧道显微镜的发展 用于检测和成像恶性病变的算法，2)STMI性能 使用真实的数字乳房模型进行评估，3)真实的物理模型 乳房幻影和第一代STMI硬件原型，以及4) 利用物理模型和体外实验评估STMI的实验性能 人体乳房组织成像场景。两类STMI算法将是 考虑：波束形成和自适应检测。波束形成器结合了 在天线阵处要通过的反向散射信号的滤波版本 从给定位置散射的信号在衰减的同时具有单位增益 来自其他方向的反向散射信号。波束形成器焦点位置是 扫描以生成作为位置函数的散射信号能量图。 自适应检测方法形成基于期望的测试统计量 如果病变实际上位于 候选地点。这两种技术都将结合频率相关 传播和散射效应与伪影的最佳鉴别 还有噪音。解剖学上逼真的数字乳房模型将被派生出来 来自常规的高分辨率3-D核磁共振乳房扫描。样机仪器 将由商业矢量网络分析器和 重新定位以创建合成阵列的超宽带天线。这个 物理乳房模型将由不同种类的固体组织模拟物组成 塑造成自然扁平的乳房的形状。体外成像 实验将在六个不同患者的组织样本上进行 从复位手术中获得的。提出的超宽带STMI方法 提供了一种无痛非电离的乳腺癌筛查的潜力 能够降低漏检率和误报率的技术 与传统的X光乳房X光检查有关，尤其是在具有挑战性的情况下 胸片致密乳房或胸壁附近肿瘤的病例 在腋下附近。

Description (Provided by Applicant): The objective of the proposed research is to demonstrate the feasibility of an ultrawideband space-time microwave imaging (STMI) method for early-stage breast cancer detection. The outcome of the proposed work will be 1) the development of robust STM about signal processing algorithms for detection and imaging of malignant lesions, 2) STMI performance evaluation using realistic numerical breast phantoms, 3) realistic physical breast phantoms and a first generation STMI hardware prototype, and 4) experimental STMI performance evaluation using physical phantoms and ex vivo human breast tissue imaging scenarios. Two classes of STMI algorithms will be considered: beamforming and adaptive detection. The beamformer combines filtered versions of the backscattered signals at an antenna array to pass signals scattered from a given location with unit gain while attenuating backscattered signals from other directions. The beamformer focus location is scanned to generate a map of scattered signal energy as a function of position. Adaptive detection methods form a test statistic based on the expected space-time pattern that would result if a lesion were actually located at the candidate location. Both techniques will incorporate frequency-dependent propagation and scattering effects and optimally discriminate against artifacts and noise. Anatomically realistic numerical breast phantoms will be derived from routine high resolution 3-D MRI breast scans. The prototype instrument will be constructed from a commercial vector network analyzer and an ultrawideband antenna that is repositioned to create a synthetic array. The physical breast phantom will consist of heterogeneous solid tissue simulants molded into the shape of a naturally flattened breast. Ex vivo imaging experiments will be conducted on six different patients' tissue samples obtained from reduction surgeries. The proposed ultrawideband STMI approach offers the potential of a pain-free non-ionizing breast cancer screening technology capable of reducing the rate of false negatives and false positives associated with conventional X-ray mammography, especially for challenging cases involving radiographically dense breasts or tumors near the chest wall or near the axilla.