Research on Disease Prediction Method Based on R-Lookahead-LSTM

Area of Science:

• Computational biology
• Medical informatics
• Machine learning in healthcare

Background:

• Cardiovascular disease poses a significant global health threat.
• High-quality disease prediction models are crucial for prevention and treatment.
• Existing models require enhancement for improved accuracy and stability.

Purpose of the Study:

• To develop a superior cardiovascular disease prediction model.
• To improve feature selection techniques for disease prediction.
• To enhance the performance and convergence of deep learning models.

Main Methods:

• Constructed novel feature vectors and analyzed their correlations.
• Employed random forest algorithm for feature importance ranking.
• Proposed a cardiovascular disease prediction model based on R-Lookahead-LSTM, utilizing Rectified Adam optimizer and Softsign activation function.

Main Results:

• Introduced three new feature vectors enhancing the original dataset.
• Random forest algorithm provided feature importance rankings.
• The R-Lookahead-LSTM model demonstrated improved stability and convergence for cardiovascular disease prediction.

Conclusions:

• The R-Lookahead-LSTM model offers a promising approach for cardiovascular disease prediction.
• Optimized algorithms and feature engineering contribute to enhanced model performance.
• This model can aid in the early detection and management of cardiovascular diseases.