An app that detects color and metadata from fingernails can identify hemoglobin levels in the blood, according to research published online Dec. 4 in Nature Communications.
Robert G. Mannino, Ph.D., from Emory University in Atlanta, and colleagues estimated hemoglobin levels by analyzing fingernail data, skin color data, and image metadata extracted from fingernail bed smartphone images. To develop the algorithm as a tool to screen for anemia, the researchers evaluated 337 participants who were split into a discovery cohort of 237 subjects and a validation group of 100 subjects. The discovery group was used to establish the correlation between image parameters and hemoglobin levels.
The researchers found that the app detected anemia with an accuracy of ±2.4 g/dL and a sensitivity of 97 percent compared with complete blood count hemoglobin levels in the validation group. With personalized calibration, the app achieved accuracy of ±0.92 g/dL of complete blood count hemoglobin levels in 16 participants.
"All other 'point-of-care' anemia detection tools require external equipment, and represent trade-offs between invasiveness, cost, and accuracy," a coauthor said in a statement. "This is a standalone app whose accuracy is on par with currently available point-of-care tests without the need to draw blood."
A patent application has been filed for the app; two authors have a financial interest in the success of the product.