Smartphone technology and the sensors that are part of smartphones or can be connected to smartphones will radically change the health sector [1]. Already today a large part of the population is using smartphones or other kinds of consumer electronics (e.g. activity trackers) to monitor their activities, their sleep, etc. Monitoring the health status of individuals with sensors is traditionally done in a clinical context, for example sleep monitoring is done with polysomnigraphy in hospital. The data collected in such studies are very important for medical doctors to make a proper diagnosis and a good plan for treatment. However, this type of studies is also very expensive and many individuals do not like hospitalization etc.
This data can nowadays also be collected at home with specialized mobile devices (e.g. the Nox T3 [2]) that are certified for clinical use. However, the devices are also very expensive. Health sensors for consumer electronics are much cheaper and for example the Bitalino [3] kit and the Coocking Hack kit [4] are also open platforms and fully programmable. However, the question is how good these sensors are and how useful their data is for diagnosis. Therefore, we aim to develop a benchmark for health sensors. In order to achieve this, we identify the following tasks
- Literature study to identify the state-of-the-art in this area.
- Analysis and evaluation of existing metrics for data quality of sensors.
- Developing solutions to measure the distance between time series data from health sensors.
- Developing a methodology and corresponding software to benchmark sensors.
Sources
- http://www.wsj.com/articles/the-future-of-medicine-is-in-your-smartphone-1420828632
- http://www.noxmedical.com/products/nox-t3-sleep-monitor
- http://www.bitalino.com
- https://www.cooking-hacks.com/ehealth-sensors-complete-kit-biometric-medical-arduino-raspberry-pi