Heart disease is considered as one of the most fatal diseases claiming millions of lives each year. To stop such diseases from growing worse, it is crucial to recognize and predict them in their early stages. Sometimes using manual processes to accurately diagnose illnesses presents a challenge for physicians. Identifying and predicting individuals at an early age of chronic heart disease are the primary goals of the proposed research. Applying different machine learning and deep learning algorithms is one way to efficiently identify individuals with chronic illnesses. Consequently, disease prediction is significantly improved when machine learning and deep learning are combined. This work presents the implementation of a unique ensemble model that combines deep learning and machine learning models for the prediction of heart disease. The novel approach is named as Deep Ensembled Machine Learning Framework (DEPMLF). The effectiveness of the DEPMLF implementation has been verified through the use of several metrics, including Confusion Matrix, Precision, Accuracy, Recall, and AUC Score. According to the findings of the experimental studies, the proposed framework performed more accurately than the traditional approaches.