Sanguina Logo

Backed By Science

At Sanguina, we pride ourselves on our transparency and commitment to the scientific process. The technology behind AnemoCheck Mobile is backed by sound clinical science. The app was developed at Emory University and the Georgia Institute of Technology using over 300 data points from individual people. The app has been tested in clinical studies at Emory University Hospital and Children’s Healthcare of Atlanta.

AnemoCheck Mobile is intended to provide you with information related to maintaining or encouraging a healthy lifestyle and is not a medical device. The information and data provided are not intended to be utilized for medical purposes and are not intended to diagnose, cure, mitigate, treat, or prevent any disease or condition. Always consult a doctor before making medical decisions.

The graphs below show how well our app works when compared to a blood test. 100 people were tested with the app. The data is presented below with blood test results on the horizontal (x) axis and the AnemoCheck app on the vertical (y) axis. The average error is 0.97 g/dL, which is roughly 10% different than a clinical blood test. 

This app should not replace a blood test.

AnemoCheck Mobile app hemoglobin (Hgb) level versus blood test (complete blood count, CBC) hemoglobin level

a) The smartphone image analysis algorithm within the AnemoCheck Mobile app measures Hgb level to within ±0.97 g/dL of the CBC Hgb level (r = 0.82, mean |error|). b) Bland-Altman analysis reveals that 95% of results fall within ±2.4 g/dL.

The graphs below show that skin tone and background lighting have little impact on hemoglobin level measurements made by the AnemoCheck Mobile app (error due to skin tone and lighting is minimal as shown in the graphs below). We recommend that the app is used with indoor lighting.

Background lighting and subject skin tone has minimal effect on AnemoCheck Mobile app accuracy.

There is little to no correlation between AnemoCheck Mobile app error and skin tone (a) or background lighting (b). (r=0.13 and 0.00, respectively). Inset images highlight a representative range of measured background skin tones and lighting conditions. n = 100 patients.

Want to learn more?

If you would like to read more about the development and validation of our app, please check out our peer-reviewed scientific journal article that we published in Nature Communications.