Anemia affects more than 2 billion people worldwide, including approximately 83 million Americans with high risk. Now, a new application offers reliable and accessible detection directly to consumers.
A new study in co -authorship of the professor at Chapman University and founding dean of the Fowler School of Engineering, Dr. L. Andrew Lyon, presents an important advance in non -invasive health technology: an application for smartphones that uses artificial intelligence and a photo of the Federna of a user to detect anemia.
Posted in the magazineProceedings of the National Academy of Sciences (Pnas)The study demonstrates that this non -invasive and augurated application of AI provides hemoglobin estimates that rival traditional laboratory tests. With more than 1.4 million tests carried out by more than 200,000 users, the application represents a low -cost scalable solution that expands access to anemia detection, especially in unattended and remote communities.
With greater access, this application brings reliable detection directly to consumers, which allows real -time health monitoring and previous intervention, allowing users to make informed decisions without waiting for laboratory results. While it is not intended for self -diagnosis, the application helps users to understand when to consult a medical care provider.
The application is particularly valuable for those with chronic anemia, such as people with kidney disease or cancer, which often require frequent monitoring. The study showed that the personalized use of the application in these patients improved accuracy in almost 50%, which allows a safer and easier handling at home.
“This research, more than eight years in creation, represents a significant step to improve accessibility in medical care,” said Dr. Lyon. “It is a testimony of long -term collaboration and a commitment to empower patients through innovation.”
Key findings:
- 4 m+ tests performed with smartphone cameras and nail analysis with AI engine.
- Hemoglobin estimates (HGB) showed an absolute half error of ± 0.72 g/dl, improving ± 0.50 g/dl in users with HGB> 10 g/dl.
- Geolocalization data enable the first prevalence map of the county anemia in the United States
- Personalization of application for patients with chronic anemia improved precision (from ± 1.36 to ± 0.74 g/dl).
- Users can now track their hemoglobin levels at home, reducing the need for frequent clinical visits.
- Traditional blood tests require a lot of time, expensive and require clinical infrastructure. This tool offers a low -cost non -invasive alternative with mass scalability.