RESUMO
Objective ToanalyzethecharacteristicmanifestationsandearlydiagnosticvalueofCTinsolitarypulmonarynodules (SPNs)lessthan2cmusing L o g istic regressionanalysis.Methods 156patientswithSPNlessthan2cmconfirmedbypathology werecollected.Statisticalassignment was performed and binary L o g istic regression wasimplemented for CT manifestations.Those features,whichmightbesignsoflungcancer,wereextractedfromtheCTimagesandtheirrisklevelswerealsoanalyzed.Results SixCTsignsincluding "ground-glasssign"(8.12),"lobulationsign"(6.72),"vascularconvergencesign"(6.02),"spiculesign"(5.07),"necrosis and cavitation "(3.41 ),and "vacuole sign (1.02 )" were enrolled in the L o g istic equation.Conclusion "Ground-glass sign "is associated with the highest risk level for lung cancer nodules.T he L o g istic m odel constructed fro m C T m anifestations is helpful for identifyingsolitarylungcancernodules.