Cardiovascular diseases pose significant health concerns, particularly especially in locations where medical facilities are limited. Smart wearable devices have emerged as a viable solution to tackle this challenge. Internet of Things (IoT) based devices serve as monitoring systems empowered by machine learning algorithms. They collect cardiovascular data, including heart rate, blood pressure, and electrocardiogram (ECG) signals. Subsequently, this data is transmitted to a centralized server for preprocessing and analysis. This paper employs a CVLSTMLW-CNN hybrid model, which combines LSTM and LWCNN. The LSTM layer analyses electrocardiogram (ECG) signals, representing the heart's electrical activity over time, and learns patterns from this sequential data. Meanwhile, the LWCNN layer extracts spatial features from medical images. By enabling early detection and intervention, the system has the potential to increase survival rates. The algorithm's efficacy is measured using metrics such as F1-score, recall, accuracy, and precision. It reliably predicts the presence or absence of cardiovascular diseases with a 99.0% accuracy across both Kaggle and UCI datasets.