Novel Coherence Imaging to Evaluate the Health of the Cervical Epithelium

描述（由申请人提供）：宫颈癌是美国发病率的重要原因，每年约有12,200例新病例和4,200例死亡。作为一种筛查工具，宫颈细胞学检查（巴氏涂片）可以通过检测癌前病变来降低发病率；但它不能定位宫颈发育不良区域，这是活检指导和治疗的必要条件。目前，怀疑发育不良的区域是通过阴道镜检查发现的，但这是一种主观诊断，准确性有限，可变性大。本研究项目的目标是开发一种新的相干成像平台，用于实际检测宫颈发育不良。达到这一目标将提高对疾病发展的诊断和理解。许多光学技术已被提出用于检测子宫颈癌前病变，但尚未广泛采用，主要是由于性能有限。虽然高分辨率方法显示出良好的灵敏度，但需要手动扫描探头穿过子宫颈外限制了实用性。相比之下，广域成像方法检查更多的组织区域，但特异性有限。在这里，我们建议创建一个基于多模态方法的成像平台，结合宽视场成像，使用多路低相干干涉测量（mLCI）进行全面的宫颈映射，使用高分辨率，深度分辨核形态学测量，使用角度分辨LCI（a/LCI）进行高度敏感的异常增生检测。本研究的长期目标是创造多模态仪器来诊断宫颈发育不良。提出以下具体目标：1)发展多路LCI用于绘制子宫颈上皮类型。mLCI探头将对宫颈外进行广域扫描，绘制鳞状到柱状的过渡区（t区）。杜克大学的原理验证研究将验证mLCI探针，随后在UCSF进行更广泛的研究。数据将与阴道摄影相比较，这是目前评估t区的标准。2)采用角度分辨LCI检测宫颈发育不良。a/LCI可以通过深度分辨核形态测量来检测发育不良病变。初步的体内试点研究将使用现有的原型进行，并将开发一种优化的仪器来扫描子宫颈外，而无需重新定位探针。该仪器将用于杜克大学的原理验证研究，然后在加州大学旧金山分校进行更广泛的研究。为了评估准确性，将a/LCI核形态学测量结果与共同登记的物理活检的组织病理学评估进行比较。3)将mLCI和a/LCI集成为单一成像平台。该集成平台将实现广域mLCI扫描，绘制子宫颈t区，并作为a/LCI核形态学测量的指南，以识别选定点的异常增生。该仪器将用于杜克大学的试点研究，然后应用于加州大学旧金山分校的临床现场测试。光学成像测量将与数字阴道摄影相比较，以验证t区映射，并与共同注册的物理活检相比较，以确认核形态与组织病理组织状态的一致性。

DESCRIPTION (provided by applicant): Cervical cancer is a significant cause of morbidity in the U.S. with approximately 12,200 new cases and 4,200 deaths from this disease each year. As a screening tool, cervical cytology (the Papanicolaou smear) can reduce morbidity by detecting precancerous lesions; but it cannot locate dysplastic regions of the cervix, a necessity for biopsy guidance and therapy. Currently, regions suspected to be dysplastic are found by colposcopy with aceto-whitening, but this is a subjective diagnostic with limited accuracy and wide variability. The goal of this research project is to develop a novel coherence imaging platform for practical detection of cervical dysplasia. Reaching this goal will improve diagnosis and understanding of disease development. Many optical techniques have been proposed for detecting pre-cancerous lesions in the cervix, but none has yet been widely adopted, due primarily to limited performance. While high resolution methods show good sensitivity, the need to manually scan a probe across the ectocervix limits utility. In comparison, wide area imaging approaches examine more tissue area, but show limited specificity. Here we propose to create an imaging platform based on a multi-modality approach, combining wide field imaging, using multiplexed low coherence interferometry (mLCI) for comprehensive cervical mapping, with high resolution, depth-resolved nuclear morphology measurements, using angle-resolved LCI(a/LCI) for highly sensitive detection of dysplasia. The long term objective of this research is to create multi-modal instrument to enable diagnosis of cervical dysplasia. The following Specific Aims are proposed. 1) Develop multiplexed LCI for mapping the epithelial types of the cervix. The mLCI probe will conduct wide area scans of the ectocervix to map the squamous-to- columnar transition zone (T-zone). A proof of principle study at Duke will validate the mLCI probe, followed by a broader study at UCSF. Data will be compared to colpophotography, the current standard for assessing the T-zone. 2) Adapt angle-resolved LCI for detection of cervical dysplasia. a/LCI can detect dysplastic lesions via depth resolved nuclear morphology measurements. An initial in vivo pilot study will be conducted using an existing prototype and an optimized instrument will be developed to scan the ectocervix without needing to reposition the probe. The instrument will be used for proof of principle studies at Duke and then validated in a broader study at UCSF. To assess accuracy, a/LCI nuclear morphology measurements will be compared to histopathological evaluation of co-registered physical biopsies. 3) Integration of mLCI and a/LCI into a single imaging platform. The integrated platform will enable wide area mLCI scans, to map the T-zone of the cervix and serve as a guide for a/LCI nuclear morphology measurements to identify dysplasia at selected points. The instrument will be used in a pilot study at Duke and then applied in a clinical field test at UCSF. The optical imaging measurements will be compared to digital colpophotography to validate the T-zone mapping and to co-registered physical biopsies to confirm alignment of nuclear morphology with histopathological tissue state.