Utilizing hybrid functional fuzzy wavelet neural networks with a teaching learning-based optimization algorithm for...

Area of Science:

• Computational intelligence
• Medical informatics
• Machine learning for healthcare

Background:

• Accurate medical disease diagnosis is a critical classification challenge.
• Existing methods may have limitations in efficiency and accuracy for complex datasets.

Purpose of the Study:

• To propose a novel hybrid classification technique for medical disease diagnosis.
• To enhance diagnostic accuracy and efficiency using an optimized neural network architecture.

Main Methods:

• A hybrid Functional Fuzzy Wavelet Neural Network (FFWNN) was developed.
• The Teaching Learning-Based Optimization (TLBO) algorithm was employed for training FFWNN and optimizing parameters (weights, dilation, translation).
• Performance was evaluated on five medical datasets (Breast Cancer, Heart Disease, Hepatitis, Pima-Indian diabetes, Appendicitis) using cross-validation.

Main Results:

• The proposed FFNN achieved high classification accuracies: 98.309% (Breast Cancer), 91.1% (Heart Disease), 91.39% (Hepatitis), 88.67% (Pima-Indian diabetes), and 93.51% (Appendicitis).
• The method demonstrated low computational complexity and efficient performance compared to existing approaches.
• Evaluation metrics included MSE, accuracy, running time, sensitivity, specificity, and kappa.

Conclusions:

• The proposed TLBO-optimized FFNN is a highly effective method for medical disease classification.
• This hybrid approach offers superior accuracy and efficiency for diverse medical diagnostic tasks.
• The study highlights the potential of combining optimization algorithms with advanced neural network models in medical informatics.