Image capture and region of interest selection. (A) Selection of conjunctiva region to be used in analysis, the square outlines the selected pixel representing best conjunctiva colour. (B) Region of interest (ROI) of the patient's palpebral conjunctiva that was selected by the ROI algorithm. Credit: Suner et al, 2021, Plos One

A new screening method using smartphone photos combined with an algorithm to detect anaemia may negate the need for an invasive blood test, reported US researchers.  

Anaemia, a low blood haemoglobin concentration, affects more than 25% of the global population and severe anaemia is a significant risk factor for morbidity and mortality, especially in children, the elderly and the chronically ill, said study lead Professor Selim Suner, Brown University, US.  

Building on prior research demonstrating that the inside of a person's lower eyelid, the palpebral conjunctiva, appears paler with anaemia, the researchers developed a study to see if smartphone imagery combined with an algorithm could provide a new inexpensive, accessible and non-invasive screening tool to diagnose anaemia.  

The researchers obtained smartphone images of the palpebral conjunctiva from 142 patients and developed an algorithm optimising colour resolution and a prediction model linking conjunctiva colour, compared to the surrounding skin and whites of the eyes, to haemoglobin levels.  

Testing the new algorithm on photos collected from 202 new patients, the model was found to be 72.6% accurate, 72.8% sensitive and 72.5% specific at predicting anaemia. The researchers found skin tone did not change results, but image quality had some effect.  

The results suggest that a smartphone app could be used to screen for anaemia in a telehealth or remote setting where the infrastructure for blood tests is not available, said Prof Suner. "Utilisation of non-invasive techniques to detect anaemia opens the door to widespread screening, early diagnosis and treatment, particularly in low resource settings where access to healthcare is sparce." 

The full study was published in Plos One.