Continuous glucose monitoring makes it possible to forecast the trajectory of future glucose concentrations. Meals, insulin, and other physiological and metabolic changes, such as physical activity, all impact glucose concentration. Devices monitoring patients’ physical activity are being developed to solve these problems. This review focuses on non-invasive sensors used to enhance glucose monitoring in patients with type 1 diabetes by utilising physiological characteristics associated with physical exercise. The search yielded 37 original research publications, from which we selected the most significant aspects regarding the devices, the various types of sensors and data acquired, the physiological signal, and the methodologies applied to analyze and use this data. The capacity to evaluate physiological data in real-time has been transformed by the growing integration of embedded artificial intelligence systems, enabling a more precise and prompt assessment of patient circumstances, including measuring glucose levels.