Biomedical engineers have developed a smartphone app for the non-invasive detection of anaemia using only patient-sourced photos. The app estimates haemoglobin levels by analysing colour and metadata of photos of the fingernail bed.
The researchers conducted a clinical assessment of the smartphone-based technology using blood samples and smartphone fingernail images of 337 patients with anaemia of different aetiologies as well as healthy subjects. The algorithm for converting fingernail colour to blood haemoglobin level was developed with 237 of these subjects and then tested on 100.
They found that a single smartphone image, without personalised calibration, could measure haemoglobin level with an accuracy of 2.4 grams/deciliter with a sensitivity of up to 97 per cent (95% CI 89%-100%). Personalised calibration, tested on four patients over the course of several weeks, improved the accuracy even further.
The developers say the app could facilitate self-management by patients with chronic anaemia, allowing them to monitor their disease and to identify the times when they need to adjust their therapies or receive transfusions, possibly reducing side effects or complications of having transfusions too early or too late.
The findings are published in Nature Communications.