Smart Watch Assisted Multi-disease Detection Using Machine Learning: A Comprehensive Survey

Due to the proliferation of wearable technologies, it has become possible to collect vast amounts of data on personal health and wellness. Smartwatches, in particular, are becoming increasingly popular due to their ability to collect data on heart rate, blood pressure, sleep patterns, and also on physical activity levels. With this data, algorithms for machine learning can be trained to predict the likelihood of developing certain diseases, such as cardiovascular disease, sleep disorders, diabetes, respiratory diseases, and neurological disorders. This review study looks at the state of the art in studies on disease prediction using data from smartwatches. We begin by exploring the types of data that can be collected by smartwatches and the machine learning algorithms used to analyze this data. Then, we reviewed research that made use of data from smartwatches to predict specific diseases and examine the accuracy of these predictions which was lying between range of 98.54% to 75.33% using machine learning algorithms such as Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and many more such. Additionally, we discuss the potential benefits and limitations of using smartwatch data for disease prediction, including privacy concerns and the need for validation studies.