Leukemia Detection using Computation Intelligence Techniques

The discipline of molecular biology is the focus of the application of statistics and computer science that is known as bioinformatics. The term "bioinformatics" refers to the study of how mathematics, statistics, and information technology might be used to the study of biological sciences. The bioinformatics of microarrays is the field’s response to the challenge posed by this definition. Genes are units of information that are responsible for encoding the genetic material found in all creatures that are alive. The term "genome" refers to the whole collection of genes that are included inside an organism. For the purpose of collecting, analyzing, storing, and exchanging genetic information with other researchers, the design of studies using microarrays is required. Because they are an organized collection of thousands of distinct Deoxyribonucleic Acid (DNA) sequences that can be used to measure variations in both DNA and Ribonucleic Acid (RNA), micro-arrays have acquired a special significance in the field of bioinformatics. This is due to the fact that micro-arrays can be used to measure both DNA and RNA. In recent years, machine learning and artificial intelligence have seen fast advancements, which has led to the methodology being one of the most widely used methods. Major corporations and educational institutions have begun investing in research in the healthcare industry with the goal of improving the accuracy with which diseases can be predicted as a result of its growing popularity and the powerful techniques it employs in pattern recognition and classification. However, while putting these strategies into practice, there are a lot of obstacles to overcome. The absence of a huge data collection including medical photographs is one of the most significant challenges that must be overcome. Because they make it possible to do quantitative research on thousands of genes all at once using a single sample of cells, microarrays in particular are being hailed as a major technological advancement in the field of biology. In the first model of an artificial neural network, two different classification algorithms—feed forward backpropagation and cas-cade correlation feedforward—were compared with different sets of neurons using two different training algorithms—Levenberg Marquardt (lm) and Resilient backpropagation (rp), in order to differentiate between benign and malignant patients. According to the findings, cascade correlation with the train (rp) produced superior output outcomes when compared to feedforward back-propagation with the train (lm).