Computer vision project to determine the quality of CPR chest compressions.
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
data
r scripts
trainer avatar
.gitignore
LICENSE
README.md
calibratedepth.py
convertdata.py
cprstatus.py
dataloader.py
datalogger.py
fixtime.py
markerfinder.py
markerutils.py
runtracker.py
statussender.py
test.py
wristtracker.py

README.md

cpr-vision-measurement

Computer vision project to determine the quality of CPR chest compressions.

A circle sticker of known size is placed on the person's wrist, facing the camera. The position of this sticker is tracked as the person performs chest compressions on a CPR dummy. This generates a 1D waveform that represents the depth and rate of the compressions. From this, the quality of the compressions can be determined (rate too fast/slow, depth too shallow/deep, partial chest recoil).

Dependencies

  • OpenCV 3.2
  • numpy 1.12.0
  • PeakUtils 1.0.3
  • (optional) matplotlib 2.0.0