American scientists have developed an algorithm that determines the level of hemoglobin in the blood from the image of the inner part of the lower eyelid of the eye. Using the application and receiving a selfie taken on a smartphone from the patient, the doctor can diagnose such a dangerous disease as anemia without additional tests. The results of the study are published in the journal PLOS ONE.
Anemia is a syndrome of a decrease in the content of hemoglobin and red blood cells in the circulating blood, which affects about one in four people in the world. In children, the elderly and chronic patients, anemia can cause severe complications and even death. This disease is especially common in developing countries, where there is a problem with its diagnosis, which usually requires complex blood tests.
Researchers from Brown University in Providence, Rhode Island, led by Selim Suner, together with colleagues from the Rhode Island Hospital, have developed a method that allows diagnosing anemia from a standard image obtained from a smartphone camera.
Previous studies have shown that the inner part of a person’s lower eyelid, called the conjunctiva of the eyelid, looks paler with anemia. In a new study, scientists obtained smartphone images of the conjunctiva of the eyeball of 142 patients with a wide range of hemoglobin levels.
They enlarged a small area of the conjunctiva in each photo and developed a color recognition algorithm, as well as a prediction model that compares the color of the conjunctiva, the surrounding skin and the whites of the eyes with the level of hemoglobin. The team then tested the created algorithm on photos collected from 202 new patients who did not participate in the initial study.
When analyzing a series of photos, the reliability of the model in predicting anemia was more than 72 percent. The patient’s skin color did not affect the sensitivity of the method in any way, and the main factor determining the accuracy of the diagnosis was the image quality.
The researchers believe that the smartphone application they developed can be used for screening anemia in telemedicine, as well as in conditions when the infrastructure for blood testing is not available.
“Images of the vascular surface of the lower eyelid obtained using a smartphone camera can be used to assess the concentration of hemoglobin in the blood and predict anemia, which is a serious disease affecting billions of people around the world, with a disproportionate effect in developing countries. The use of non-invasive methods for detecting anemia opens the door to widespread screening, early diagnosis and treatment, especially in countries with limited resources, where access to healthcare is limited,” the authors write.