The healthcare industry is advancing rapidly in both technology and services. One recent development is remote health monitoring, which has become increasingly important in a world where the aging population is facing more health complications. Initially, this technology was limited to monitoring patients within hospital rooms. However, advancements in communication and sensor technologies have made it possible to monitor patients while they go about their daily activities at home. One popular device being used for this purpose is the smartwatch, due to its efficiency and ease of use in transmitting health data quickly and conveniently via smartphones. This study proposes an end-to-end remote monitoring framework for predicting and managing health risks using different types of personal health devices, smartphones, and smartwatches. Several machine learning methods were applied to a collected dataset, which underwent feature scaling, imputation, selection, and augmentation to predict health risks. The tenfold stratified crossvalidation method achieved an accuracy of 99.5%, a recall of 99.5%, and an F1 of 99.5%, which is competitive with existing methods. Patients can utilize various personal health devices, such as smartphones and smartwatches, to monitor vital signs and manage the development of their health metrics, all while staying connected with medical experts. The proposed framework allows medical professionals to make informed decisions based on the latest health risk predictions and lifestyle insights while maintaining unobtrusiveness, reducing cost, and ensuring vendor interoperability. The cost of entire system is 328 USD.