New Approch to Red Blood Cell Classification Using Morphological Image Processing

lranian journal of Clinical Infectious Diseases

Behzad Baradaran¹  Alireza Mehdizade³  Mojtaba Taherisadr³  Mona Nasirzonouzi² 

[1] Research Center of Immunology, Tabriz University of Medical Sciences, Tabriz, IR Iran;Azad University of Tabriz, Tabriz, IR Iran;Research Center of Biomedical Physics and Engineering, Shiraz University of Medical Sciences, Shiraz, IR Iran

关键词: Automatic classification;    Red blood cells;    Image processing;    Blood disorders;   

DOI  :  10.17795/semj18147

学科分类：基础医学

来源: Shiraz University of Medical Sciences

【 摘 要 】

Blooddisorderscancausemorphologicalchangesinmatureredbloodcellssobyinvestigatingbloodsmearsmorphologically,wecanstudyanddistinguishblooddiseasessuchasanemias.Ontheotherhand,becausemostoftheavailablemethodsaremanual,expensive,time-consuminganddependontheexpertsexpertisesoforimprovingdetectionrateandincreasingitsaccuracy,amethodofautomatedanalysisfortherapidclassificationoflargenumbersofredcellsfromindividualspecimenshasbeendeveloped.Accordingtothismethodwhichisbasedondigitalimageprocessing,severalfeaturesrelatingtoshape,internalcentralpallorconfigurationofredcellsandtheircircularityandelongationhavebeenextractedandwiththehelpofadecisionlogicallthosevarioustypesofredbloodcellswereclassifiedinto12categories.Accordingtotheobtainedresults,diagnosisofblooddisorderssuchasirondeficiencyanemia,theanemiaofchronicdisease,β-thalassemiatrait,sicklecellanemia,hemoglobinCdisease,intravascularhemolysis,hereditaryelliptocytosis,hereditaryspherocytosisandmegaloblasticanemiaduetofolicaciddeficiencycanbepossible.

【 授权许可】

Unknown