Research has found that photographing the inner lower eyelid with a smartphone can be used to detect anemia.

Recently, U.S. scientists published a new study in the journal PLOS ONE, demonstrating that by capturing images of the inner eyelid with a smartphone and applying an algorithm to analyze these photos, they can estimate blood hemoglobin levels and thereby screen for anemia. The method achieves an accuracy of 72%. This means that in the future, individuals will be able to easily monitor their own anemic status, with people in remote areas, in particular, able to use a mobile app to self‑assess whether they are anemic.

According to the Medical Information Network, approximately 25% of the global population suffers from anemia. Individuals with anemia have hemoglobin levels below the normal range, and common symptoms include fatigue, dizziness, headache, shortness of breath, and difficulty concentrating. Severe anemia is primarily caused by malnutrition, parasitic infections, and other factors; it can be life‑threatening and is more prevalent among vulnerable populations such as children, the elderly, and patients with chronic diseases.

Medical professionals typically use sophisticated laboratory equipment to measure hemoglobin levels in the blood to diagnose anemia. However, in many remote areas, medical infrastructure is inadequate, leaving local residents unable to promptly determine whether they are anemic or to assess the severity of their condition. Dr. Selim Suner, a researcher at Brown University and Rhode Island Hospital, explains that once diagnosed with anemia, patients can improve their condition by taking iron supplements: “The treatment is inexpensive and convenient, but the real challenge lies in accurately diagnosing anemia in the first place.”

Accordingly, Dr. Selim Suner is working to develop an inexpensive, convenient, and noninvasive screening method—using smartphone photography to detect anemia. To evaluate the feasibility of using a smartphone camera for anemia detection, Dr. Selim Suner and his team conducted a two-phase study.

Phase 1: In the emergency department, mobile phone photographs of the inner eyelids were taken for 142 patients.

Researchers found that the color contrast between the blood vessels of the inner eyelid and the conjunctival surface is pronounced, the distance between them is very small, and they are minimally influenced by environmental factors. Consequently, the researchers selected the inner eyelid as the subject of their imaging, magnified each photograph of the inner eyelid, and employed algorithms to maximize color resolution for predictive analysis, thereby comparing hemoglobin levels in the ocular skin with normal reference values.

Phase 2: The algorithm was tested on mobile phone images from 202 patients in the emergency department.

The results indicate that the model achieves an accuracy of 72.6% in predicting anemia. For patients with severe anemia requiring blood transfusion, the prediction accuracy is even higher, ranging from 86% to 94.4%.

The study demonstrates that using mobile phones to predict anemia is a viable concept. Dr. Selim Suner stated, “This represents a particularly compelling opportunity for developing countries, where healthcare systems may be sparse, rudimentary, and unevenly distributed. This project could have a positive impact on large populations and contribute to global health.”

However, the researchers note that using a smartphone to detect anemia still has certain limitations, such as variations in image quality, since test subjects may pull their eyelids closed during image capture. Furthermore, the study did not specify the lighting conditions under which images were taken, making it unclear whether changes in illumination might affect image quality.