Cardiovascular Disease Prediction Based on an Improved Dendritic Neuron Model

Cardiovascular diseases (CVDs) are not only an epidemic but also have a high mortality rate. Predicting the onset of CVDs in advance can significantly reduce mortality rates. However, researchers have found it difficult to quickly and accurately diagnose CVDs. In this study, we employed the whale genetic algorithm (WGA) to train an improved dendritic neuron model (DNM) for CVD prediction. Specifically, due to the powerful optimization performance of the WGA, the dendritic neuron model can converge quickly and yield a high classification performance. Three benchmark datasets related to CVDs are used to verify the proposed WGA-DNM. We compare WGA with five optimization algorithms to verify whether it can serve as an effective training method. In addition, we also compare the proposed WGA-DNM with five classical classifiers to assess its classification capabilities. The experimental outcomes show that the proposed WGA-DNM performs well based on various metrics and can provide decision support for the diagnosis of CVDs.