A Diagnostic Mannequin for Acute Lymphoblastic Leukemia Utilizing Metaheuristics and Deep Studying Strategies
Authors: M. Hosseinzadeh, P. Khoshaght, S. Sadeghi, P. Asghari, Z. Arabi, J. Lansky, P. Budinsky, A. Masoud Rahmani, S. W. Lee
Summary: Acute lymphoblastic leukemia (ALL) severity is set by the presence and ratios of blast cells (irregular white blood cells) in each bone marrow and peripheral blood. Handbook analysis of this illness is a tedious and time-consuming operation, making it troublesome for professionals to precisely study blast cell traits. To deal with this issue, researchers use deep studying and machine studying. On this paper, a ResNet-based function extractor is utilized to detect ALL, together with a wide range of function selectors and classifiers. To get the very best outcomes, a wide range of switch studying fashions, together with the Resnet, VGG, EfficientNet, and DensNet households, are used as deep function extractors. Following extraction, totally different function selectors are used, together with Genetic algorithm, PCA, ANOVA, Random Forest, Univariate, Mutual info, Lasso, XGB, Variance, and Binary ant colony. After function qualification, a wide range of classifiers are used, with MLP outperforming the others. The beneficial method is used to categorize ALL and HEM within the chosen dataset which is C-NMC 2019. This method acquired a formidable 90.71% accuracy and 95.76% sensitivity for the related classifications, and its metrics on this dataset outperformed others
