Classification of Leukemia Blood Samples Using Neural Networks

Abstract

Pattern recognition applied to blood samples for diagnosing leukemia remains an extremely difficult task which frequently leads to misclassification errors due in large part to the inherent problem of data overlap. A novel artificial neural network (ANN) algorithm is proposed for optimizing the classification of multidimensional data, focusing on acute leukemia samples. The programming tool established around the ANN architecture focuses on the classification of normal vs. abnormal blood samples, namely acute lymphocytic leukemia (ALL) and acute myeloid leukemia (AML). There were 220 blood samples considered with 60 abnormal samples and 160 normal samples. The algorithm produced very high sensitivity results that improved up to 96.67% in ALL classification with increased data set size. With this type of accuracy, this programming tool provides information to medical doctors in the form of diagnostic references for the specific disease states that are considered for this study. The results obtained prove that a neural network classifier can perform remarkably well for this type of flow-cytometry data. Even more significant is the fact that experimental evaluations in the testing phase reveal that as the ALL data considered is gradually increased from small to large data sets, the more accurate are the classification results.

Artificial neural network in breast lesions from fine‐needle aspiration cytology smear

“…1,2 Recently, ANNs have been used in various fields such as facial recognition, credit card fraud detection, handwriting recognition, and various fields of medicine such as radiological image interpretation, survival prediction, and cytological diagnosis. [3][4][5][6][7][8][9][10][11][12][13] However, to the best of our knowledge, various cytological features along with quantitative image morphometry have never been used in building up an ANN model for breast cancer detection. In this study, we extracted the cytological features and morphometric data of fine-needle aspiration cytology (FNAC) of ductal carcinomas of breast and fibroadenomas and built an ANN for diagnosis of benign and malignant cases.…”

Genetic Selection of Fuzzy Model for Acute Leukemia Classification

“…Adjouadi et al [1] use data collected with a flux cytometer and a neural network is implemented for the recognition of acute leukemia, which is designed to recognize between normal cells or abnormal cells (ALL and AML). The data set has 220 samples (160 are normal cells and 60 are abnormal cells).…”

“…The coding scheme adopted is described in detail next. The feature selection method is represented by two bits: a value of [0, 0] means information gain; [0, 1] represents feature selection based on correlation, a value of [1,0] Finally, the learning algorithms are represented by a variable number of bits, according to the parameters required by each of them. This means that the proposed genetic algorithm works with individuals of different size.…”

Predicting First-Episode Psychosis Associated with Cannabis Use with Artificial Neural Networks and Deep Learning

“…On the one hand, artificial neural networks have been successfully used in understanding the heterogeneous manifestations of asthma [7], diagnosing tuberculosis [8], classifying leukaemia [9], detecting heart conditions in ECG data [10], etc. These studies show that neural networks have been proven to be capable of dealing with complicated medical data such as the ambiguous nature of the ECG signal data, where neural networks show some outstanding results compared to other methods.…”