Acute Lymphoblastic Leukemia Cell Detection in Microscopic Digital Images Based on Shape and Texture Features

• Published in: Advances in Swarm Intelligence Vol. 11656; pp. 142 - 151
• Main Authors: Tuba, Eva, Strumberger, Ivana, Bacanin, Nebojsa, Zivkovic, Dejan, Tuba, Milan
• Format: Book Chapter
• Language: English
• Published: Switzerland Springer International Publishing AG 2019; Springer International Publishing
• Series: Lecture Notes in Computer Science
• Subjects: Acute lymphoblastic leukemia detection; Bare bone fireworks algorithm; Local binary pattern; Optimization; Segmentation; Support vector machine; Swarm intelligence
• ISBN: 3030263533; 9783030263539
• ISSN: 0302-9743; 1611-3349
• DOI: 10.1007/978-3-030-26354-6_14

Leukemia or blood cancer is a disease that affects a large population, especially children. Fast and early detection of four main types of leukemia is crucial for successful treatment and patient’s recovery. Leukemia can be detected in microscope blood images by detecting blasts, i.e. not fully developed white blood cells. Computer-aided diagnostic systems can improve the quality and speed of abnormal lymphocytes detection. In this paper we proposed a method for automatic detection of one type of leukemia, acute lymphoblastic leukemia, by classifying white blood cells into normal cells and blasts. The proposed method uses shape and texture features as input vector for support vector machine optimized by bare bones fireworks algorithm. Based on the results obtained on the standard benchmark set, ALL-IDB, our proposed method shows a competitive accuracy of classification comparing to other state-of-the-art method.