Ambient Healthcare Approach with Hybrid Whale Optimization Algorithm and Naïve Bayes Classifier

Area of Science:

• Health Informatics
• Data Science
• Computational Intelligence

Background:

• Healthcare systems face challenges processing large, complex patient data efficiently.
• Existing cloud-based IoT healthcare systems struggle with processing time and overall efficiency for big data.

Purpose of the Study:

• To develop a novel, computationally efficient approach for processing big healthcare data.
• To improve data classification accuracy and reduce processing time for immediate decision-making.
• To enhance healthcare business agility, security, privacy, and reduce operational costs.

Main Methods:

• A hybrid algorithm combining Whale Optimization Algorithm (WOA) for feature selection and Naïve Bayes Classifier (NBC) for real-time classification.
• Implementation leveraging Fog Computing architecture for distributed data processing.
• Evaluation of the approach on datasets for Diabetes, Heart Disease, Heart Attack Prediction, and Sonar.

Main Results:

• Significant reduction in the number of dataset features through WOA.
• Improved classification accuracy by an average of 3.6% across tested datasets.
• Reduced processing time by an average of 8.7%, enhancing real-time capabilities.

Conclusions:

• The proposed hybrid WOA-NBC approach effectively processes big healthcare data with reduced computational cost.
• The method offers a promising solution for improving efficiency and accuracy in healthcare data analytics.
• Fog Computing integration enhances the system's agility, security, and privacy while lowering operational expenses.