Innovations in cervical cancer diagnosis for low resource settings using advanced optical imaging and machine learning diagnostic algorithms.

The broad goal of this project is to adapt a portablelow costeasy to use Pocket sized Colposcopedeveloped under other fundingfor use in a community settingand develop automated algorithms that combine neovascularizationglycogen depletion and acetowhitening to provide comparable diagnosis to an expertThis work will be done in a collaboration betweenrd Stone DesignIncDuke University and Kenya Medical Research InstituteThe specific aims of this proposal areAimPhase IImprove Pocket colposcope by designing continuous magnification mechanism and improving device workflow integration to eliminate between use disinfection through the use of a disposable optically clear sterile sleeveProvider feedback on our previously developed Pocket colposcope has unanimously suggested the addition of a slider mechanism to control coarse zoom and a sleeve consumable to the Pocket colposcope designAimPhase IAutomated algorithms and software for cervical pre cancer detection We will improve the specificity of VIA using a novel software application with embedded machine learning diagnostic algorithms for automated cervical cancer screeningWe will apply and validate the individual algorithms for VIA and GIVIgreen illumination vascular imagingto existing images obtained from apatient clinical study with the Pocket colposcopeWe will then compare the performance of the algorithms to expert physician interpretation of the same imageswith pathology serving as the gold standardAimPhase IIDocument user experience with Pocket colposcope in KenyaWe will develop a culturally relevant training package directly in the community healthcare settingWe will collect quantitative and qualitative data including surveysin depth interviewsand clinic observations from both naive providers and patients and use these findings to and use these findings to improve the introduction of the Pocket colposcope in Kenya and simultaneouslyinform the clinical investigations in AimAimPhase IICompare the performance of the Pocket colposcope to Visual Inspection with Acetic Acid for triage of HPVwomen in KenyaWe will carry out a cluster randomized trial amongHPVwomen to compare the standard triage with that using the Pocket colposcope in KisumuKenyaAll HPVwomen will undergo biopsy to determine sensitivityspecificity and positive and negative predictive values of the different triage strategiesData will be used to model the performance of the algorithm against that of expert colposcopistsAimPhase IIAssess the costsincremental cost effectiveness and population health impact of HPV based cervical cancer screening programs with proposed triage strategiesWe will determine the incremental cost effectiveness ratio and the absolute and relative costs for four triage strategies by measuring the costs and model population health outcomescancer casesdeaths and disability adjusted life yearsNARRATIVE The SBIR activities proposed byrd Stone Design and Duke University will advance the state of the art in cervical cancer imaging through the Randamp D of novel optics for a portable colposcope combined with automated algorithms for diagnosisThe project will confirm the benefits of these innovations through a clinical trial in Kenya conducted by the Kenya Medical Research InstitutesThe innovations developed have tremendous potential to effect public health by increasing diagnostic acuity and decreasing costs of healthcare delivery which ultimately will reduce the impact of the deadly disease