Open AccessClassifying White Blood Cells in Blood Smear Images using a Convolutional Neural Network
Publication year - 2019
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.i1133.0789s19
Subject(s) - convolutional neural network , binary classification , artificial intelligence , white blood cell , peripheral blood , pattern recognition (psychology) , class (philosophy) , computer science , peripheral blood mononuclear cell , binary number , blood smear , artificial neural network , white (mutation) , multiclass classification , blood cell , medicine , support vector machine , immunology , biology , mathematics , biochemistry , arithmetic , malaria , gene , in vitro
We have tried to automate the classification task of white blood cells by using a Convolutional Neural Network. We have divided white blood cell classification in two types of problems, a binary class problem and a 4-classification problem. In binary class problem we classify white blood cell as either mononuclear or Grenrecules. In 4-classification problem where cells are classified into their subtypes (monocytes, lymphocytes, neutrophils, basophils and eosinophils). In our experiment we were able to achieve validation accuracy of 100% in binary classification and 98.40 in multiple classifications.