A Computer Tool for Aiding in Accurate Assessment of Indeterminate Lung Nodules

Low dose CT screeningLDCThas been approved by the Centers for Medicare andampMedicaid ServicesCMSfor Medicare coverageHoweverthe primary problem associated with CT is the high false positivedetectionsThis issue often leads to unpleasant and costly unintended consequencese gfollow up scans and or invasive biopsiesIn this projectwe propose to develop and commercializea novel computer tool to aid in accurate assessment of indeterminate lung nodulesThe goal is toaccurately and efficiently quantify the potential risk of developing lung cancer and its futureprognosisthereby facilitating precisepersonalized lung cancer screening and optimal patientmanagementDuring Phase I of this projectwe have accomplished the proposed milestones anddeveloped a prototype system that is now publicly accessibleOur preliminary validation of theprototype system demonstrates very promising performanceIn Phase IIwe will continue our effortto make this tool available to serve clinical community with the following specific aimsdevelop a generalized framework that supports the incorporation of both image and patientinformation as well as other biological tests related to lung cancerfully optimize thecomputer algorithms to efficiently analyze chest CT scans and in particular synergize ouralgorithms with the deep learning technology to improve training efficiency and benign malignanceclassification accuracyandcomprehensively validate and optimize the system in clinicalenvironment at the University of Pittsburgh Medical CenterUPMCWe believe that the proposedsystem is extremely timely and important in light of the CMS decision to cover annual LDCT lungcancer screeningIts availability will significantly reduce the over diagnosis associated withLDCT for lung cancer screening and relieve the economic burden on healthcare system Project NarrativeThe project aims to develop an online computer tool to aid in efficacious earlylung cancer screening and diagnosis based on LDCTThis tool will not only locate the pulmonarynodules automaticallybut also quantitatively assess the malignancy of a suspicious nodule and itsprognosis