Smartphone app that can accurately extract information about a person’s blood hemoglobin content

Smartphone app that can accurately extract information about a person’s blood hemoglobin content from a photo of the inner eyelid has been developed by a  Purdue associate professor of biomedical engineering, and his team and collaborators. Using a smartphone-taken photo of the inner eyelid, the advanced learning algorithm implements spectral super-resolution spectroscopy and further analyzes microvascular and blood hemoglobin information.

Our team has developed an mHealth algorithm that reliably estimates blood hemoglobin level from a photo of the inner eyelid. Blood hemoglobin tests routinely are ordered as an initial screening to check general health status before other specific examinations, and are one of the most common clinical lab tests. But conventional blood testing via a blood draw is not ideal: It’s invasive, takes time to produce a result, and can cause excessive blood loss when multiple tests are necessary.

Blood hemoglobin tests also are extensively performed to assess hematologic disorders, transfusion initiation, hemorrhage detection after traumatic injury, and acute kidney injury. As an example, anemia, a major public health problem in developing countries, is characterized by low levels of blood hemoglobin. The World Health Organization (WHO) has estimated that anemia affects 24.8 percent of the global population — more than 1.6 billion people. Anemia is particularly prevalent in Africa, affecting two-thirds of preschool-age children and half of women. Importantly, malnutrition and nutritional anemia — the latter caused when the body does not absorb sufficient amounts of certain nutrients — are two sides of the same coin.

We envision several uses for our mHealth technology. It is practical in emergency room and ICU settings when repeated blood hemoglobin measurements are necessary, avoiding excessive blood loss. It’s a useful tool in resource-limited and home care settings that lack access to centralized clinical laboratories. And as malnutrition and nutritional anemia are so closely entwined, it can empower local healthcare workers in rural villages — enhancing malnutrition awareness and advancing healthcare in low- and middle-income countries.

Our mHealth innovation is based on super-resolution spectroscopy that “virtually” transforms the built-in camera of a smartphone into a hyperspectral imager for accurate and precise blood hemoglobin analyses. “Super-resolution” means high-resolution reconstruction of digital signals acquired with low-resolution systems. Conventional methods require bulky and costly optical components; we overcome these limitations through statistical and deep-learning computational algorithms that help us to analyze photos of the microvasculature redness of the inner eyelid. Our research illustrates how a data science approach can minimize hardware complexity, facilitating effective scale-up and translation into commercial availability and use.

We’re developing a user-friendly mobile app as the “face” of the algorithm. We will conduct a clinical review with a full usability test, including analysis of intra- and inter-blood hemoglobin measurements. As part of the process, we have initiated a clinical study of 5,000 patients in collaboration with Shrimad Rajchandra Hospital in Dharampur, India, which serves a largely rural population in the south of Gujarat, a state on the western coast of India.

We were awarded first prize in a National Institutes of Health (NIH) Technology Accelerator Challenge to design and develop non-invasive, handheld, digital technologies to detect and diagnose anemia, sickle cell disease, and malaria. The National Institute of Biomedical Imaging and Bioengineering (NIBIB) led the competition, with a mission to support the development of new diagnostic technologies that are crucial for global health.

Connected digital tools like our mHealth app are vital to extending the continuum of healthcare to dramatically enhance patient care, disease surveillance, decision and policy making, and clinical outcomes — leading to improved health for individuals and populations as a whole.