Optimizing Resource Utilization of Big Data in Healthcare for Cardiovascular Disease Prediction

Big data in healthcare involves gathering, analysing, and interpreting vast amounts of medical information, such as electronic health records (EHR), diagnostic images, and real-time sensor data from wearable devices. With the growing volume of healthcare data, there is an increasing demand for sophisticated machine learning models that can process this data and extract meaningful insights. One of the most important uses of healthcare big data is predicting cardiac diseases, where early and accurate diagnoses can significantly enhance patient outcomes. Long Short-Term Memory (LSTM) networks with attention mechanisms, known for their ability to process sequential time-series data, are particularly effective in analyzing dynamic signals like ECGs and heart rate data. The attention mechanism improves the LSTM's ability to focus on the most relevant parts of the input data, thus boosting model performance, especially in handling complex sequences. This study introduces an LSTM with Attention Mechanism model that surpasses traditional machine learning algorithms in classifying cardiac diseases, yielding higher precision and recall. These findings highlight the value of LSTM with attention mechanisms in enhancing the accuracy, efficiency, and interpretability of predictive models for cardiovascular diseases, providing a scalable solution for real-time, data-driven healthcare systems.