An Improved Long Short-Term Memory Algorithm for Cardiovascular Disease Prediction.

Area of Science:

• Biomedical Engineering
• Artificial Intelligence in Medicine
• Cardiovascular Health

Background:

• Cardiovascular diseases are a leading cause of mortality, necessitating early and accurate diagnostic methods.
• Existing diagnostic models face challenges in network configuration and performance, limiting their accuracy.
• Effective early diagnosis is crucial for cardiovascular disease risk prevention.

Purpose of the Study:

• To introduce a novel, robust model for the early diagnosis of cardiovascular diseases.
• To enhance the accuracy and efficiency of cardiovascular disease diagnostic models.
• To address limitations in network configuration and performance degradation in current models.

Main Methods:

• Development of the Optimally Configured and Improved Long Short-Term Memory (OCI-LSTM) model.
• Utilization of the Salp Swarm Algorithm for irrelevant feature elimination.
• Application of the Genetic Algorithm for optimizing LSTM network configuration.

Main Results:

• The OCI-LSTM model demonstrated high efficacy, validated by accuracy, sensitivity, specificity, and F1 score.
• Comparative analysis showed the OCI-LSTM model's superiority over Deep Neural Networks and Deep Belief Networks.
• A significant accuracy increase of 97.11% was achieved with the OCI-LSTM model.

Conclusions:

• The OCI-LSTM model presents a promising advancement for accurate and efficient early diagnosis of cardiovascular diseases.
• The model's robust feature selection and optimized network configuration contribute to its superior performance.
• Future research should focus on real-world clinical implementation and further refinement of the OCI-LSTM model.