Hybridization of Artificial Gravity Cuckoo Search Algorithm with XGboost-Particle Swarm Optimized Neural Networks for Cardiac Feature Selection

One of the worst and most prevalent human illnesses, heart disease kills 4 in 10 people worldwide and affects 80–90% of individuals in the United States, Wales, Canada, and the United Kingdom. Numerous techniques exist for the early diagnosis of cardiac disease; however, due to the enormous amount of data, few pass quality control. Effective smart strategies for the early prediction of heart disease have been presented to address these study limitations. In this study, we enhance the workflow of an automated heart disease detection system using the artificial gravity cuckoo search algorithm and XGboost-Particle Bee optimized neural network. Based on how cuckoos hatch, the approach chooses heart traits. The suggested strategy increases the overall accuracy of heart disease prediction by routinely updating the weight and bias parameters. XGboost PSOWNN algorithm has the best accuracy 99.96% than other method BAT-BP, GA-CNN, ACONN. Finally, we use MATLAB to assess the system’s effectiveness. This lowers the total mistake rate and increases the precision of recognition.