Improved Swarm Intelligence Optimization Model with Mutual Information Estimation for Feature Selection in Microarray Bioinformatics Datasets for Diseases Diagnosis

Bioinformatics has emerged as a promising field with innovative applications in various bio-logical domains. Microarray data analysis has become the preferred technology for estimating gene expression and diagnosing diseases. However, processing and computing the vast amount of information contained in microarray data pose significant challenges. This study focuses on developing an effective feature selection model for disease diagnosis using microarray datasets. The proposed approach, mutual fuzzy swarm optimization (MFSO), uses mutual information (MI) values to compute features in a microarray gene dataset. The computed MI values are then applied to a fuzzy expert system (FES) for sample classification. To improve classification accuracy, fuzzy logic is iteratively estimated for unclassified instances in the microarray sample. The feature selection model employs a particle swarm optimization model to extract microarray sample features based on the derived MI. The swarm optimization move-ment is guided by an objective function. The expert system integrates fuzzy logic with human expert knowledge to perform medical diagnoses, with a specific focus on the selected features. The proposed MFSO model utilizes if-then rules to estimate the membership values of the microarray dataset features. Through simulation analysis, the performance of the proposed MFSO model is evaluated using sensitivity, specificity, and ROC values for five different microarray datasets. The results demonstrate improved performance compared to existing methods. In conclusion, this study presents a novel approach for effective feature selection in a microarray dataset for disease diagnosis. The proposed MFSO model integrates MI computation, fuzzy logic, and expert knowledge to achieve improved classification accuracy. The simulation analysis validates the effectiveness of the proposed model, highlighting its superior performance compared to existing methods.