Researchers from Purdue University, Rwanda Biomedical Center, and the University of Rwanda used common smartphones to take more than 12,000 eye photos of 565 children between the ages of 5 and 15. They then used machine learning in conjunction with radionics, a method that uses mathematical analysis of patterns and textures in medical images to find characteristics associated with anemia. The journal Biophotonics Discovery published this study.
Radiomic features extracted through machine learning from black-and-white smartphone photos enable noninvasive anemia detection in school-age children. This mobile health approach can potentially be valuable in remote or resource-limited settings. The original demonstration photos performed by the authors are further rendered using ChatGPT. Image Credit: Y. L. Kim, S. G. Hong, and Haripriya Sakthivel
Nearly 2 billion people worldwide suffer from anemia, a disorder characterized by low hemoglobin levels in the blood.
If treatment is not received, children's anemia can impede their general development, learning, and growth. Early detection is crucial, but standard diagnostic techniques require lab equipment and blood samples, which are frequently unavailable in low-income areas.
The study used straightforward grayscale images of the conjunctiva, the white portion of the eye, and the inner surface of the eyelid to forecast anemia.
Unlike previous efforts that rely on color analysis or special imaging tools, this method does not require color data. Instead, it uses black-and-white photos to examine tiny structural changes in the eye’s blood vessels. This approach avoids problems caused by different light conditions or camera models, making it more practical for use in a variety of settings.
Shaun Hong, PhD Student and Study First Author, Purdue University
The findings indicate that there may be a way to screen for anemia with just a smartphone and simple software because they demonstrate a strong correlation between certain spatial characteristics and anemia status.
This could be particularly helpful in isolated or underdeveloped areas, providing a quick, non-invasive, and reasonably priced method of identifying children who are at risk.
The technology is not meant to replace traditional testing but could help prioritize who needs further evaluation and treatment. With more development, the method could be integrated into mobile health tools to support early intervention in areas where healthcare access is limited.
Young L. Kim, Study Corresponding Author and Professor, Purdue University
Journal Reference:
Hong, S. G., et al. (2025) Radiomic identification of anemia features in monochromatic conjunctiva photographs in school-age children. Biophotonics Discovery. doi.org/10.1117/1.bios.2.2.022303.