IEEE Consumer Electronics Magazine - September/October 2020 - 46

Healthcare Cyber-Physical System

with recommendations. Wagner et al.10 conducted a study on 113 patients asking them to
measure their blood pressure according to the
mentioned guidelines. Patients used a system
that can verify if recommended conditions were
met. The authors concluded that none of them
fully respected the recommendations. With the
increasing number and diversity of connected
sensors, it is now possible to detect situations
that are not in line with recommendations. Consequently, these sensors have the potential to
help patients perform a reliable blood pressure
measurement.
In this section, we discuss how connected
sensors can help in verifying recommendations
mentioned in Table 2, as they are the most common ones, then how such a system can be implemented practically with regard to the actual
market situation. First of all, a blood pressure
measurement should be done at normal temperature (Rec. 1). It is possible to verify this condition using home automation sensors such as the
Fibaro Motion Sensor, or a connected weather
station such as the Netatmo one. It is a sensor
composed of a main indoor module, with the
possibility of multiple extensions, and one outdoor module. The main module communicates
over Wi-Fi. Blood pressure should also be measured in a quiet room (Rec. 2), which can be verified with any microphone-based device. The
third recommendation is to be at rest during the
measurement (Rec. 3) and not to exercise within
30 minutes prior to the measurement (Rec. 4).
Smartwatches, wristbands, and actimeters are
accelerometer-based sensors that are able to
determine the physical activity level of their
users, and therefore to detect if they were
exercising or not before the measurement. Examples of these devices are the iHealth Wave, Fitbit
Charge, and Apple Watch. Regarding smoking
(Rec. 5), connected cigarette cases or electronic
cigarettes can tell when users smoked for the
last time. Also, Akyazi et al. developed an application that runs on smartwatches and detects
any smoking action with an accuracy of 96%,16
provided that the smartwatch is worn on the
same arm as that used for smoking. To detect alcohol (Rec. 6) and caffeinated beverages
(Rec. 7), a smart bottle should be used. Smart
bottles are devices able to analyze their contents

in order to identify what the user is currently
drinking. Until recently, such a device was still
under development. Regarding the user's posture (Rec. 8), sensors such as UpRight Go can
tell if their users are slouching or not. As for
detecting crossed legs (Rec. 9), no sensor has
been commercialized so far.
By combining these types of sensors, it is possible to guide users measuring their blood pressure while respecting recommended conditions
to perform this measurement. For a practical
implementation of such a system, it is important
to take into consideration the current market situation. Temperature can be measured using connected thermostats, increasingly installed in
homes. Noise recommendations can be verified
using smartphone or tablet devices. As for physical activity sensors, they are widely used. Gartner
estimated that smart wristband shipments would
reach 48.84 million units in 2018. Regarding smoking sensors, they are either in a very early step
of development, or not popularized (around
100 downloads of the Smoking Stopper application according to Google Play). Concerning
smart bottles, Vessyl manufacturer closed down.
Consequently, no sensor on the market is able to
verify alcohol or coffee conditions. The smartphone application of the UpRight Go sensor cumulated more than 10 000 downloads on Google
Play. The company also raised over one million
dollars through the crowdfunding platform Kickstarter. This indicates that users are willing to
adopt this kind of device.
Thus, in regard to the current market situation, only temperature, noise, physical activity,
and straight back recommendations can be easily
verified using commercialized sensors. Accordingly, Figure 3 shows how a context-aware blood
pressure measurement can be completed. It illustrates for each step of the flowchart a few types
of sensors that can be used to perform the
measurement.

Privacy
In the context of health and medicine, data
privacy is a fundamental pillar ensuring that
patients' data remain confidential. With the
increasing number of connected sensors, privacy
in home health care, and more generally in telemonitoring systems, is becoming a real challenge

IEEE Consumer Electronics Magazine

IEEE Consumer Electronics Magazine - September/October 2020

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - September/October 2020