In developing countries, anemia is a major public health problem including children and pregnant women. In Africa, the prevalence of anemia is extremely high, affecting two thirds of preschool-age children and a half of women. When anemia is not detected and managed in a timely manner, it can result in major health consequences, including fatigue, heart failure, pregnancy disorders and poor physical/cognitive conditions. Early and accurate diagnosis of anemia can reduce a need for complicated treatments. Although portable (in-house) or point-of-care blood analyzers are currently available, these still remain unaffordable and inadequate in low- and middle-income country settings. Clinical examinations, digital photography, and smartphone-based noninvasive measurements have also received attention, but these still lack sensitivity and specificity for anemia diagnosis. Considering the imperative clinical need in developing countries, the `right' technology would be an accurate and sensitive noninvasive hemoglobin measuring technology that can easily be embedded into conventional mobile smartphones. We will develop a mobile smartphone-based noninvasive spectrometer-less hemoglobin analyzer (mHema) and to conduct studies with Academic Model Providing Access to Healthcare (AMPATH) program in Kenya. Our approach takes advantage of virtual hyperspectral (each pixel has detailed color information) imaging to have similar results compared with laboratory-based hemoglobin tests (gold standard). This platform will be implemented as a user-friendly application in conventional mobile smartphones. We propose the following three specific aims: i) develop a two-step algorithm for accurate and sensitive hemoglobin quantification, ii) implement the hemoglobin detection algorithm within an electronic health record system- integrated smartphone application, and iii) evaluate the performance characteristics of mHema in a real-world clinical setting in Kenya. As the proposed study is at the intersection of medical device development, clinical laboratory testing, and computer science, our interdisciplinary team is well poised to carry out this proposed work as a part of AMPATH. Given that the detection technology is completely noninvasive and rapid, we will piggyback patients undergoing routine blood tests. Successful completion of the planned studies will facilitate clinical translation of this mobile application tool and maximize the quality of current available standard of care. This work will provide a foundation to comprehensively evaluate the accuracy and precision of mHema for a larger patient population with a broader set of conditions, further serving as a model clinical setting to tailor our technology for other resource-limited settings.

Public Health Relevance

Anemia, which is defined as a decrease in the amount of hemoglobin in the blood, is a major public health problem in low- and middle-income countries. However, portable/point-of-care hemoglobin measurement modalities often remain inaccessible/unaffordable and current noninvasive mobile-based technologies still lack sensitivity/specificity for different levels of anemia. In this respect, we will develop mobile smartphone-based noninvasive hemoglobin analyzer (mHema), implement a user-friendly application in conventional mobile smartphones, and evaluate the accuracy and precision of mHema in low-resource settings.

Agency
National Institute of Health (NIH)
Institute
Fogarty International Center (FIC)
Type
Exploratory/Developmental Grants (R21)
Project #
5R21TW010620-02
Application #
9567216
Study Section
Special Emphasis Panel (ZRG1)
Program Officer
Povlich, Laura
Project Start
2017-09-18
Project End
2019-08-31
Budget Start
2018-09-01
Budget End
2019-08-31
Support Year
2
Fiscal Year
2018
Total Cost
Indirect Cost
Name
Purdue University
Department
Engineering (All Types)
Type
Biomed Engr/Col Engr/Engr Sta
DUNS #
072051394
City
West Lafayette
State
IN
Country
United States
Zip Code
47907
Kim, Taehoon; Visbal-Onufrak, Michelle A; Konger, Raymond L et al. (2017) Data-driven imaging of tissue inflammation using RGB-based hyperspectral reconstruction toward personal monitoring of dermatologic health. Biomed Opt Express 8:5282-5296