Testing Global Histogram Equalization and Unsharp Mask Algo-rithms for Processing Conventional Chest X-Ray Images.

[摘要] Introduction::Imagingmethodsareprogressinginarapidlymanner,buttheproblemwhichwe,asthehealthprovidersalwaysencounterwithistheexpensivecostsofdifferentdevicesandourlimitedbudgettoprovidethem.Aims::TheaimofthisstudyistoevaluatetheusefulnessofHistogramEqualization(HE)andUnsharpMask(UM)ontheconventionalCXRimages.MethodsandMaterial::InUrmiaUniversityofMedicalSciences,wedesignedawindows-basedcomputerprogramthatcontainshistogramequalization(HE),unsharpmask(UM)andcom-binationofHEandUMalgorithmswithadjustedparameterstoprocessconventionalchestx-ray(CXR)images.TwoseriesofCXRimagesincluding49imageswithoutmajorpulmonarydisorderand45imageswithpulmonaryparenchymaldisorderswereselected.Afterconvert-ingthemtodigitalformat,imageswereprocessedwithHE,UMandcombinationofHEandUMtechniques.Ineachseries,originalandprocessedimagesweresavedin4databases.Twoboard-certifiedgeneralradiologists(with6and5yearsexperience)analyzedimages.Savedimagesweredisplayedtoradiologistsrandomlyandseparately.Qualityofeachimagewassavedasascalefrom1(verylowquality)to5(excellent).Weusedavariance-basedstatisticaltechniquetoanalyzequality.Statisticalanalysisused::Tocomparethequalityofeachalgorithm(GHE,UMandcombina-tionofGHEandUM),avariance-basedstatisticalanalysiswasdone.Results::Inthefirstseriesimages,HEandcombinationofHEandUMalgorithmsincreasedqualityofimages,butUMtechniquewasnotsuitable,solely.Also,allthreetechniquesin-creasedqualityofsecondseriesimages.Conclusions::TheuseofdigitalimageprocessingalgorithmssuchasHEorUMonconven-tionalCXRimagescanincreasequalityofimages.

[学科分类] 基础医学

[关键词] Image Processing;enhancement;Chest;Conventional;Radiography