Imaging CAncer Screening Patch (CASP) for early diagnostics of cervical cancers

点击翻译按钮获取中文摘要

Imaging CAncer Screening Patch (CASP) for early diagnostics of cervical cancers The work pursued within this proposal will carry out the feasibility studies to develop a new class of high resolution, non-invasive, optical multi-field of view, flexible surface imaging devices for cervical cancer detection and screening. The device is a response for an unmet crucial need of a lack of point-of-care diagnostic tools for cervical dysplasia and cancer in patients with an abnormal screening Pap and/or HPV test. The proposed tool called Cancer Screening Patch or CASP will be capable of providing pathology level resolution over large areas of tissue and therefore can become an effective diagnostic instrument. To simplify the screening procedure the system will be merged with image analysis tools to provide objective parameters like nuclear to cytoplasmic ratio (NCR) and allow screening by low/medium training level personnel. The initial results will be available within minutes, allowing for a “Screen & Treat” approach and eliminating the need for pathology review and associated wait times and additional return visits. The core idea of the approach is combining high resolution micro-endoscopes into multi-point flexible imaging patch array conforming to the shape of cervix and providing high sampling map available for numerical analysis. Specifically we will develop prototypes of CASP capable of sub-cellular imaging resolution over large tissue areas with image sampling of 2-4 microns at 100-150 FOVs. The target individual probing areas will have diameter of 200-450 microns and will be densely populated over imaged tissue and converged into a single camera output. In parallel we will develop effective analysis tools to provide objective assessment parameters like nuclear to cytoplasmic ratio, NCR thresholding and border delineation. To further increase imaged area and resolution, the probe can be linearly or rotationally shifted and imaged zones will be merged with mosaicking algorithms. Finally we will evaluate system imaging performance using both tissue models as well as ex-vivo LEEP specimens from 20 patients with a high- grade dysplasia in a pilot clinical study.