An Efficient Strategy for Blood Diseases Detection Based on Grey Wolf Optimization as Feature Selection and Machine Learning Techniques

Abstract

Acute Lymphoblastic Leukemia (ALL) is a cancer that infects the blood cells causing the development of lymphocytes in large numbers. Diagnostic tests are costly and very time-consuming. It is important to diagnose ALL using Peripheral Blood Smear (PBS) images, especially in the initial screening cases. Several issues affect the examination process such as diagnostic error, symptoms, and nonspecific nature signs of ALL. Therefore, the objective of this study is to enforce machine-learning classifiers in the detection of Acute Lymphoblastic Leukemia as benign or malignant after using the grey wolf optimization algorithm in feature selection. The images have been enhanced by using an adaptive threshold to improve the contrast and remove errors. The model is based on grey wolf optimization technology which has been developed for feature reduction. Finally, acute lymphoblastic leukemia has been classified into benign and malignant using K-nearest neighbors (KNN), support vector machine (SVM), naïve Bayes (NB), and random forest (RF) classifiers. The best accuracy, sensitivity, and specificity of this model were 99.69%, 99.5%, and 99%, respectively, after using the grey wolf optimization algorithm in feature selection. To ensure the effectiveness of the proposed model, comparative results with other classification techniques have been included.

Statistical performance review on diagnosis of leukemia, glaucoma and diabetes mellitus using AI

“…From Table 1 , we can infer the following: Statistical performance of DL algorithms 5,11 NN + Bayesian Regularization and UNET+AlexNet GoogleNet+ SqueezeNet outperforms than the ML algorithms 5 such as, SVM, NB, LOG, KNN, DT for “ALL‐IDB2” dataset. The input dataset used by few research works 1‐4 are very less in number. Grey Wolf optimization based ML algorithm 4 too performs equally well for other input dataset. DL algorithms DCNN (binary) and EfficientNet performs well than other algorithms The statistical performance of a particular ML/DL algorithm behaves differently for different input dataset, and system design (training, testing data etc.) in a study.…”

“…• Statistical performance of DL algorithms 5,11 NN + Bayesian Regularization and UNET+AlexNet GoogleNet+ SqueezeNet outperforms than the ML algorithms 5 such as, SVM, NB, LOG, KNN, DT for "ALL-IDB2" dataset. • The input dataset used by few research works [1][2][3][4] are very less in number.…”

Hybrid Techniques for the Diagnosis of Acute Lymphoblastic Leukemia Based on Fusion of CNN Features

“…The deviation appears in t(12;21)/ETV6-RUNX1 blast cells with high CD10 and CD34 and low CD81 expression. Nada et al [ 27 ] used a machine-learning algorithm for acute leukemia classification based on a feature-selection algorithm by the gray-wolf optimization method. Adaptive thresholding was applied to improve images and then classify them by SVM, KNN, and NB.…”

Application of Machine-Learning Algorithms for Better Understanding the Properties of Liquisolid Systems Prepared with Three Mesoporous Silica Based Carriers

“…Machine learning (ML) algorithms have been widely used in various areas of health sciences in the last two decades. Their ability to process large amounts of multidimensional data and interpret the mutual influence of a large number of variables means that they can be used for more efficient analysis of medical data and early diagnosis [ 28 , 29 , 30 ]. The application of ML in drug delivery has become increasingly prominent with the development of more complex techniques such as nanosystems or various approaches to personalized medicine, including 3D printing [ 31 , 32 ].…”