AbstrakProsedur tradisional klasifikasi sel darah menggunakan mikroskop di laboratorium hematologi dilakukan untuk memperoleh informasi jenis sel darah. Telah menjadi landasan di laboratorium hematologi untuk mendiagnosis dan memantau gangguan hematologi. Namun, prosedur manual melalui serangkaian uji laboratorium dapat memakan waktu cukup lama. Oleh karena itu penelitian ini ditujukan khusus untuk dapat membantu dalam proses tahap awal klasifikasi jenis sel darah putih secara otomatis di bidang medis.Upaya untuk mengatasi lamanya waktu dan untuk keperluan diagnosis awal dapat menggunakan teknik pengolahan citra berdasarkan morfologi sel darah. Penelitian ini bertujuan untuk mengklasifikasi sel darah putih berdasarkan morfologi sel dengan k-nearest neighbor (knn). Algoritma pengolahan citra yang digunakan adalah hough circle, thresholding, ekstraksi ciri. kemudian untuk proses klasifikasi digunakan metode k-nearest neighbor (knn).Pada proses pengujian digunakan 100 citra untuk di ketahui jenisnya. Hasil pengujian segmentasi menunjukkan akurasi sebesar 78 % dan pengujian klasifikasi sebesar 64%.
Kata kunci-Sel darah putih, hough circle, ekstraksi ciri, k-nearest neighbor (k-nn)
AbstractThe traditional procedure of classification of blood cells using a microscope in the laboratory of hematology to obtain information types of blood cells. It has become a cornerstone in the laboratory of hematology to diagnose and monitor hematologic disorders. However, the manual procedure through a series of labory test can take a while. Thresfore, this research can be helpful in the early stages of the classification of white blood cells automatically in the medical field.Efforts to overcome the length of time and for the purposes of early diagnose can use the image processing technique based on morphology of blood cells. This research aims to classify the white blood cells based on cell morphology with the k-nearest neighbor (knn). Image processing algorithms used hough circle, thresholding, feature extraction, then to the process of classification was used the method of k-nearest neighbor (knn).In the process of testing used 100 images to be aware of its kind. The test results showed segmentation accuracy of 78% and testing the classification of 64%.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.