DIASCOPIC, LLC — Department of Health and Human Services SBIR Phase I: NIBIB

Description

Abstract Summary In this SBIRwe propose to validate our handcrafted image analysis algorithm for auto detecting Mycobacterium tuberculosisMTBin a digitized sputum smearOnce validated in a blinded study against manual microscopy and culturethe gold standardwe will try to improve our handcrafted algorithm by integratingwhere appropriatedeep learning approachesvia Convolutional Neural NetworksCNNOur novel diagnostic devicethe Diascopic iON platformuses automated image analysis to detect pathogens of interestThrough a blinded studyslideswe will assess the iONandapos s effectiveness in detecting MTBOur aim is to achieve andgtaccuracy vsmicroscopyand sensitivity specificity vsculture ofandrespectivelyCurrentlythe iON platform can detect MTB on a Ziehl NeelsenZNstained sputum smear in less thansecondswith accuracy ofvsmicroscopyThe primary objective of this SBIR is to meet or exceed the minimal requirements for the WHO Target Product Profilepublishedof a rapid sputum based test for detecting TB at the microscopy center level of the health care systemWe will accomplish this feasibility study through a collaborative effort with the Case Western Reserve University UgandaUCRCresearch teamA full slide digitization and automated image analysis ofZN slides is planned while on the ground in UgandaResults will be published in an appropriate peer reviewed journal for dissemination to the relevant TB pathology and provider communityA secondary objective of this SBIR is to improve our handcrafted algorithm through the use of deeplearning techniquesCNNWe will collaborate with DrMadabhushiCase Western Reservea world leader in Deep Learning methodologieson this portion of the studyWe are optimistic that by combining our handcrafted approach with a deep learning approachwe can identify MTB bacilli more effectivelyi efaster and more accuratelyWe will leverage the lessons learned in this study to develop algorithms for other developing world diseases like Onchocercariver blindnessPlasmodiummalariaand ShistomesschistosomaisisSuccessful completion of this SBIR will show that the iON can truly become a platform for automated pathogen detectionwhich will shift lab practices toward faster andampmore standardized routines that are performed by unskilled workersIf weandapos re successful in this Phase I SBIRwe will develop auto detect algorithms forother pathogens in a phase II SBIRWe will then market the iON platform to resource limited clinics in countries adversely affected by developing world diseasesIt is our experience that such clinics are seeking a rapidlow costaccurate and simple diagnostic tool to improve their efficiency and their ability to detect and treat diseases Narrative This SBIR is a validation study of a digital pathology platform to detect TB in digitized Ziehl NeelsenZNslidesWe aim to establish a high accuracyandgtvsmanual microscopy and a sensitivity andampspecificity ofandrespectivelyvscultureThe TB analysis occurs rapidlywith results available in andltsecondsWe will investigate whether algorithm improvements are possible by combining our handcrafted approach with deep learning approaches to improve accuracy and efficiencyIf high accuracy and sensitivity specificity can be achieved for TB detectionthis low cost technology can have a significant impact on TB laboratory operations around the worldThe technology can also be applied to other pathogens whose primary method of detection is microscopy