Related Experiment Video
Updated: Sep 22, 2025
07:35
Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances
Published on: October 11, 2018
7.7K
Feature Subset Selection with Optimal Adaptive Neuro-Fuzzy Systems for Bioinformatics Gene Expression Classification.
Anwer Mustafa Hilal1, Areej A Malibari2, Marwa Obayya3
1Department of Computer and Self Development, Preparatory Year Deanship, Prince Sattam Bin Abdulaziz University, AlKharj, Saudi Arabia.
Computational Intelligence and Neuroscience
|May 24, 2022
Summary
This study introduces a novel feature selection method using an optimal adaptive neuro-fuzzy inference system (OANFIS) for gene expression classification. The approach achieved up to 89.47% accuracy on cancer datasets.
Area of Science:
- Bioinformatics and computational biology
- Artificial intelligence in medicine
- Fuzzy systems applications
Background:
- Fuzzy systems are increasingly utilized in bioinformatics for applications like gene expression analysis and medical data classification.
- Developing effective microarray gene expression classification models requires advanced computational techniques.
- Artificial intelligence and fuzzy systems offer promising solutions for complex biological data analysis.
Purpose of the Study:
- To introduce a novel feature subset selection with an optimal adaptive neuro-fuzzy inference system (FSS-OANFIS) for enhanced gene expression classification.
- To detect and classify gene expression data accurately using an optimized computational model.
- To improve the performance of gene expression classification through advanced feature selection and parameter tuning.
Main Methods:
- A novel FSS-OANFIS model was developed, incorporating an improved grey wolf optimizer-based feature selection (IGWO-FS) for optimal feature subset derivation.
- The OANFIS model was employed for gene classification, with its parameters tuned using the coyote optimization algorithm (COA).
- The model was validated on Leukemia, Prostate, DLBCL Stanford, and Colon Cancer gene expression datasets.
Main Results:
- The proposed FSS-OANFIS model demonstrated effective gene expression classification capabilities.
- The integration of IGWO-FS and COA techniques led to enhanced classification outcomes.
- The model achieved a maximum classification accuracy of 89.47% across the tested datasets.
Conclusions:
- The FSS-OANFIS model presents a robust approach for accurate gene expression classification.
- The study highlights the effectiveness of combining fuzzy systems with AI optimization algorithms for biological data analysis.
- The proposed method shows significant potential for applications in cancer research and diagnostics.
