Nature-inspired optimization techniques for cardiovascular disease detection: a comprehensive survey

One of the most common illnesses that can shorten a person’s life span nowadays is heart disease. The early detection of aberrant heart diseases is crucial for identifying heart issues and preventing sudden cardiac death. Cardiovascular diseases, commonly referred to as heart diseases, are brought on by bad lifestyle choices including smoking, drinking alcohol, and eating a lot of fats, which can led to diabetes, hypertension, and other conditions. They include coronary heart disease and are brought on by conditions of the heart and blood arteries (heart attacks). The benefit of using optimization methods for complicated nonlinear situations is their adaptability and flexibility. This research provides a concise analysis of the most researched combination of various optimization techniques and machine learning techniques for cardiac disease prediction currently available in the literature along with the variations of several nature-inspired algorithms. These algorithms are also compared with regard to convergence, accuracy, feature reduction, and other aspects of nature-inspired optimization. The last 10 years’ worth of heart disease datasets are also thoroughly reviewed. This paper’s objective is to review various optimization methods like particle swarm optimization (PSO), ant colony optimization (ACO), artificial bee colony (ABC), spider monkey optimization (SMO), and many more nature-inspired optimization methods are discussed for effective cardiac disease diagnosis. The results of the investigation demonstrate that the accuracy rates for logistic regression with particle swarm optimization (PSO) is 98.33% and 65% accuracy, when k-closest neighbor (k-NN) is utilized with particle swarm optimization (PSO). These numbers are high when compared to the current decision tree, naive Bayes, and random forest approaches. According to that analysis, the Cleveland heart disease dataset is particularly well-liked by academics because it includes less missing values. It is clear from the thorough research and comparison that particle swarm optimization (PSO) will deliver superior results in comparison with other techniques currently being used for the detection of cardiac problems.