Efficient deep spectral logistic decision neural network (DSLDNN) for cardiovascular disease stratification: Evaluating performance in python and MATLAB

Globally, the prevalence of cardiovascular diseases (CVD) has increased significantly across all age groups, and the disease’s death rate is also increased. Early detection of the symptoms associated with cardiovascular disease is often difficult to do, which increases many people’s chances of dying young. As a result, providing early diagnosis and accurate disease classification will help patients seek appropriate treatment for the issues related to cardiovascular disease and significantly improve lifespan. Numerous machine learning algorithms are implemented in the current research, and the methods’ effectiveness is examined by using them with a range of research instruments. In order to classify cardiovascular disease, the proposed method focuses on building a novel machine learning classification algorithm called the Deep Spectral Logistic Decision Neural Network (DSLDNN). Proposed algorithm’s performance is compared with five pre-existing classifiers, including Support Vector Machine (SVM), Random Forest (RF), Decision Tree (DT), Ada Boost, and Gradient Boost in Python and MATLAB. The simulations are obtained successfully in both Python and MATLAB, and the outcomes clearly show that although both systems can execute DSLDNN efficiently. MATLAB provides the highest level of accuracy not only for the current algorithms but also outlines a significant variance of classifier precision for the new DSLDNN algorithm.