The cases of heart disease continue to increase across the globe and, thus, necessitate the introduction of new proactive diagnosis and management. The current practice of using wearable IoT devices and advanced machine learning models has become the immediate need for preventive measures. This work meets this need by exploring which combination of the available machine learning algorithms will best predict heart disease such as Artificial Neural Network (ANN), Long Short-Term Memory (LSTM) networks, Decision Tree (DT), and Support Vector Machine (SVM). In this research, an experiment has been carried out by deploying sensors for monitoring heart rate, blood pressure, ECG signals, pulse oximetry, temperature, and respiration, which culminated in the data generation of 3200 records. Preprocessing of the data was done to filter out relevant features from the data, and consequently, these were used to train and test machine learning models. The ANN achieved the top performance with an accuracy of 98.99% and an AUC-ROC value of 0.99, stating high potency in data pattern recognition. The LSTM model was also well performed with an accuracy of 95.6% and an AUC-ROC value of 0.96, capable enough for time-series data from ECG signals. The results can be used to develop highly advanced diagnostic tools with wearable technology for continuous health monitoring for the early and accurate detection of heart diseases.