Sensor playground: implementation and evaluation of a tool for the selection of fit-for-purpose wearable sensors in sport studies.
Mayr Sebastian (2022): Sensor playground: implementation and evaluation of a tool for the selection of fit-for-purpose wearable sensors in sport studies.
During the selection of an appropriate wearable sensor for measuring human biomarkers in sports studies, many considerations have to be made and there exist several obstacles that make this process difficult. One of these factors is the huge abundance of different devices to choose from, each possessing
a multitude of properties that need to be assessed in order to choose a suitable device. In addition, often a disorganized selection approach is applied. Therefore, a structured process is needed that provides practical assistance with this critical aspect of study design.
A scoping review was performed to collect information on how the sensor selection is performed throughout various scientific disciplines. The results were synthesized into an independent framework, which was implemented as a web application and evaluated by domain experts for its usability using the System Usability Scale (SUS).
The literature analysis identified factors that were considered as important in many sensor selection frameworks. The main factors turned out to be pragmatic. Ease of use and similar sensor-specific requirements throughout the disciplines were preferred. These requirements constitute the base for the implementation of a web application, which we call Sensor Playground.
The evaluation, done by 7 sports scientists with experience in the selection of wearable sensors, provided positive qualitative feedback. Furthermore, System Usability Scale was applied to quantify the usability of the devised system in a range from 0 to 100. A value of 68.9 indicates its acceptability, although improvements were suggested.
These findings show that experts are aware of the difficulties in searching for appropriate devices and there is a real need for a sensor selection support tool. The evaluation showed that they are willing to accept such a tool for their own work. Nonetheless, there is still room for improvement in terms of design, operation, and data quality.
