An efficient and low complex model for optimal RBM features with weighted score-based ensemble multi-disease...

Area of Science:

• Artificial Intelligence
• Computational Biology
• Medical Informatics

Background:

• Current disease prediction models often focus on single diseases, failing to address the common reality of patients suffering from multiple conditions simultaneously.
• Accurate identification of multiple diseases at various progression stages during clinical visits is crucial for effective patient management.

Purpose of the Study:

• To develop and implement a novel multi-disease prediction model leveraging ensemble learning and deep features.
• To enhance the accuracy of simultaneous disease detection by optimizing feature extraction and prediction algorithms.

Main Methods:

• Utilized a Deep Belief Network (DBN) for initial feature extraction from RBM layers.
• Optimized extracted features using Deviation-based Hybrid Grasshopper Barnacles Mating Optimization (D-HGBMO).
• Implemented an ensemble learning model combining Deep Neural Network (DNN), Extreme Learning Machine (ELM), and Long Short-Term Memory (LSTM) with optimized features for prediction.

Main Results:

• The proposed ensemble model demonstrated effective multi-disease prediction capabilities.
• Optimized features using D-HGBMO significantly improved prediction performance.
• Experimental results confirmed the superiority of the proposed model compared to existing classifiers based on quantitative measures.

Conclusions:

• The developed multi-disease prediction model offers a promising approach for identifying multiple concurrent diseases.
• The integration of deep features, optimization techniques, and ensemble learning enhances predictive accuracy in complex clinical scenarios.
• This method provides a valuable tool for clinicians to simultaneously detect various diseases, improving diagnostic efficiency.