Hello #medtwitter! Join me in my second-ever #Tweetorial re:
Can smartphone apps measure blood pulse oximetry reliably?
@Marty_Fried @OhioStateIMRes @cjchiu @AvrahamCooperMD @ErinMichos
@willcjensen @APNDMD @DrTomasGuerrero @JoeySalibaMD @andrewjohnsmd
Can smartphone apps measure blood pulse oximetry reliably?
@Marty_Fried @OhioStateIMRes @cjchiu @AvrahamCooperMD @ErinMichos
@willcjensen @APNDMD @DrTomasGuerrero @JoeySalibaMD @andrewjohnsmd
I& #39;ve been following wearable technology for several years now. The new Apple Watch Series 6 and iOS 14 is rumored to feature a pulse oximeter support and has gained some serious hype in the setting of #Covid_19.
But first, what exactly is pulse oximetry?
But first, what exactly is pulse oximetry?
Pulse oximetry is a non-invasive method used to measure the peripheral oxygen saturation (SpO2) using hemoglobin (Hb) saturation. It& #39;s a close approximation of arterial oxygen saturation (SaO2), but not always (for example, hypoperfusion, CO/CN toxicity, methemoglobinemia).
Deoxygenated & oxygenated Hb absorb light at different wavelengths (660 nm and 940 nm, respectively).
Deoxygenated Hb allows more infrared light to pass through & absorbs more red light. Oxygenated Hb absorbs more infrared light & allows more red light to pass though.
Deoxygenated Hb allows more infrared light to pass through & absorbs more red light. Oxygenated Hb absorbs more infrared light & allows more red light to pass though.
Transmissive (more common) measurement sends two wavelengths of light through an ear lobe or fingertip to the photodetector.
Reflectance (less common) measurement can use the wrist or forehead but less accurate due to:
• vasodilation
• venous pooling
• environmental light
Reflectance (less common) measurement can use the wrist or forehead but less accurate due to:
• vasodilation
• venous pooling
• environmental light
Samsung Galaxy phones had a red light emitting diode (LED) built into the phone in addition to the flash light/camera module as far back as 2014. The app was then subsequently pulled in 2019 and relocated to a "measure stress" function within another app.
Note to iPhone users that you have only a camera white-LED light and do not even have this additional red LED. This makes any app that assures you the capability to measure SpO2 a little more suspect.
This article published in The Verge yesterday addresses this topic a little further... https://www.theverge.com/2020/4/23/21232488/blood-oxygen-apps-iphone-samsung-unreliable-fitbit-garmin-oximeter">https://www.theverge.com/2020/4/23...
It references the Centre for EBM (The University of Oxford) article linked below written by Professors Tarassenko (electrical engineering) and Greenhalgh (primary care medicine). *Note: it is not formally peer-reviewed. https://www.cebm.net/covid-19/question-should-smartphone-apps-be-used-as-oximeters-answer-no/">https://www.cebm.net/covid-19/...
The listed lit-review article concludes:
• Not physically possible to measure SpO2 using current smartphone technology.
• The two published studies which assessed smartphone oximeter apps (Digidoc and Samsung) raise serious questions about the diagnostic accuracy.
• Not physically possible to measure SpO2 using current smartphone technology.
• The two published studies which assessed smartphone oximeter apps (Digidoc and Samsung) raise serious questions about the diagnostic accuracy.
A few more articles linked here for your own review, if interested:
http://rc.rcjournal.com/content/64/Suppl_10/3238714
https://rc.rcjournal.com/content/6... href=" https://www.ncbi.nlm.nih.gov/pubmed/29215972
https://www.ncbi.nlm.nih.gov/pubmed/29... href=" https://www.ncbi.nlm.nih.gov/pubmed/31471076
https://www.ncbi.nlm.nih.gov/pubmed/31... href=" https://www.ncbi.nlm.nih.gov/pubmed/30904343 ">https://www.ncbi.nlm.nih.gov/pubmed/30...
http://rc.rcjournal.com/content/64/Suppl_10/3238714
https://rc.rcjournal.com/content/6... href=" https://www.ncbi.nlm.nih.gov/pubmed/29215972
https://www.ncbi.nlm.nih.gov/pubmed/29... href=" https://www.ncbi.nlm.nih.gov/pubmed/31471076
https://www.ncbi.nlm.nih.gov/pubmed/31... href=" https://www.ncbi.nlm.nih.gov/pubmed/30904343 ">https://www.ncbi.nlm.nih.gov/pubmed/30...
Aside from smart phone apps, other consumer wearable companies have rolled out wrist-based SpO2 sensors (FitBit, Garmin, etc.) and integrated this data into their native apps. You can also find stand-alone pulse oximeters beginning at ~$15-25. https://www.wareable.com/wearable-tech/pulse-oximeter-explained-fitbit-garmin-wearables-340">https://www.wareable.com/wearable-...
It& #39;s interesting to see the trend consumer wearables set in the healthcare-technology market. Ingenious software is still limited by hardware capabilities and physiological limitations.
Being able to obtain a sufficient waveform with minimal artifact is also important.
Being able to obtain a sufficient waveform with minimal artifact is also important.
Conclusion: No. Camera-based apps are not reliable.
At best, the results were mildly inconsistent in patients without true hypoxia. In the setting of hypoxia, results were even more skewed.
Not to mention, these apps are not FDA cleared for medical use.
At best, the results were mildly inconsistent in patients without true hypoxia. In the setting of hypoxia, results were even more skewed.
Not to mention, these apps are not FDA cleared for medical use.