Harnessing Optimized Progressive Dropout Dense Networks for Accurate Heart Disease Prediction in Healthcare Systems

Abstract

The field of medical sciences has seen substantial diversification due to developments in computational capabilities and methodologies, namely in diagnosing cardiovascular disorders in humans. Cardiovascular disease (CVD) is a highly challenging condition with significant adverse effects on the global population, greatly impacting human well-being. The prompt and precise detection of cardiovascular diseases in persons can provide substantial advantages in the management of the progression of heart failure during its initial phases, hence enhancing the likelihood of the patient's life. The identification of heart illness by manual processes is prone to prejudice and vulnerable to variations among examiners. Machine learning algorithms have demonstrated efficacy and reliability in identifying and classifying persons with heart illness and those in a healthy state. A novel approach, the Optimised Progressing Dropout Dense Network (OPDDN), has been devised to identify heart illness accurately. By meticulously adjusting and refining particular parameters, we achieved an exceptional accuracy level, reaching an impressive rate of 99%. This accomplishment is deserving of recognition. In this study, a comparison was conducted between the proposed strategy and an alternative method, which led to the observation that our method demonstrated a higher level of effectiveness. This innovative methodology has the potential to function as a more efficient tool for medical practitioners in the diagnosis of cardiovascular conditions. The findings of our study suggest that implementing this innovative approach has the potential to augment the advancement of more efficacious treatment methodologies for persons diagnosed with heart disease.