Utilization of <i>K</i>-nearest neighbor algorithm for classification of white blood cells in AML M4, M5, and M7

Abstract Acute myeloid leukemia (AML) M4, M5, and M7 are subtypes of derived from cell derivatives that influences the results identification AMLs, which includes myeloblast, monoblast, megakaryoblast. Furthermore, they divided into more specific types, including myeloblasts, promyelocytes, monoblasts, promonocytes, monocytes, megakaryoblasts, must be clearly identified in order to further calculate ratio value blood. Therefore, this research aims classify these types using K -nearest neighbor (KNN) algorithm. Three distance metrics tested, namely, Euclidean, Chebychev, Minkowski, both weighted unweighted were tested. The features used as parameters area, nucleus ratio, circularity, perimeter, mean, standard deviation, about 1,450 objects training testing data. In addition, ensure classification is not overfitting, K- fold cross validation was conducted. show Minkowski acquired 240 290 at = 19, best. selected for analysis. accuracy, recall, precision values KNN with obtained fivefold 80.552, 44.145, 42.592%, respectively.