Page 162 - ITU Kaleidoscope 2016

establishing a voice communication path between a patient- We applied this system to the stand-up and sit-down
PC and an operator-PC. After a patient selects an operator, training, as shown in Figure 11. Both a patient’s display and
the video call center server matches the operator_PeerID an operator’s one when a patient is doing stand-up and sit-
with the patient_PeerID. The SkyWay client_libraries on down training are shown in Figure 12. We present strain of
each PC establish a voice communication path between a the upper body by using three view formats on an operator’s
patient-PC and an operator-PC in combination with the display: the front view, the side view, and the top view.
SkyWay server. An operator selects a view format and clicks a button to
measure data. These data are temporally stored as a file.
4.3. Measuring tool After an operator decides to save them on the server, the file
is stored as an archived file. This sequence flow is shown in
After data and voice paths are established between a Figure 13. An operator can also bookmark selected frames.
patient and an operator, the operator requests the patient to
start the Microsoft-KINECT v2 application so that the
operator can monitor the patient and data can be measured.
This sequence flow is shown in Figure 6.

Figure 8. Flow diagram for aggregating video images
and skeleton data

Figure 6. Sequence flow to start Microsoft KINECT

Since video images and skeleton data are independently
output from Microsoft KINECT, both have to be
synchronized to enable monitoring at an operator’s office as
shown in Figure 7. We thus serialize skeleton data in the
JSON format and add them to each item of video frame data, Figure 9. Flow diagram for composing pictures
which is resized, transferred to the JPEG format, and then
transferred to the Base 64 text data at the application on a
patient’s PC, as shown in Figure 8. These data are sent to a
Web server in the video call center. They are separated
again and individually drawn on the canvas of the html
monitoring page, as shown in Figure 9. Video image data
are drawn on a lower layer and skeleton data are drawn in
our presentation format on a transparent upper layer, as
shown in Figure 10.

Figure 10. Example of composing video image and skeleton
data

Figure 7. Monitoring scheme of patient’s motion

– 144 –