Rat Swarm Optimization and Machine Learning Algorithms for Accurate Leukemia Diagnosis via Gene Expression Analysis

Cancer can be caused by a sharp rise in the number of cells that might damage internal organs. Leukemia is a form of cancer that affects the body's blood-forming tissues, including the lymphatic system and bone marrow. Recent technical developments have made it possible to quickly assess the cellular makeup of various cancer forms. Each properly positioned tiny dot on a microscope slide printed with a DNA microarray contains either a DNA sequence or a gene. This study aims to enhance machine learning algorithms like Passive Aggressive Algorithm, Ridge Classifier, Quadratic Discriminant Analysis, XG Boost Algorithm, Stochastic Gradient Descent algorithm, K Nearest Neighbour Classifier, Random Forest Classifier, and Perceptron Model by using Rat Swarm Optimization as a transform. In this analysis, gene expression data from 282 patients with leukemia was taken into account. To reduce the number of dimensions in a dataset, Principal Component Analysis (PCA) is used. Importantly, Ridge Classifier delivers the greatest Balanced Accuracy score of 94.92% for Leukemia cancer data when combined with Rat Swarm Optimization.