Prophecy of Cardiac Diseases with XGBoost and Gray Wolf Algorithm

Treating cardiac illness, a widespread health concern, requires a prompt and precise diagnosis. Machine learning techniques for medical diagnosis problems, particularly ensemble algorithms such as XGBoost, have demonstrated encouraging outcomes. To enhance the performance of these models hyperparameters tuning is required. This study enhances the diagnosis of cardiac illness by combining the XGBoost algorithm with the Gray Wolf Search Algorithm (GWSA). To optimise XGBoost classifier hyperparameters such as regularisation, tree depth, and learning rate GWSA is utilized. The research was conducted using a large set of clinical and diagnostic data from patients with different heart problems. Data preprocessing made sure that scaling was consistent and handled missing data. When combined XGBoost with GWSA, improves the accuracy of algorithm for cardiac problems more than when using traditional parameter tuning techniques. Numerous metrics demonstrate the enhanced XGBoost model's ability to distinguish between different heart states. The outcome of proposed model shows accuracy 97.8% which is significantly higher than traditional ML algorithms. The proposed model have precision 97%, recall 89% and F1-score 93%. Explanations of the interpretability of the model and the significance of the features for diagnostic decision-making are explained in paper. The accuracy and reliability of heart disease detection may be raised by using XGBoost and swarm intelligence algorithms such as GWSA. The suggested techniques in clinical settings enhance patient care and healthcare results.