Abstract
BackgroundIn many low and middle-income countries (LMICs), difficulties in patient identification are a major obstacle to the delivery of longitudinal care. In absence of unique identifiers, biometrics have emerged as an attractive solution to the identification problem. We developed an mHealth App for subject identification using pattern recognition around ear morphology (Project SEARCH (Scanning EARS for Child Health). Early field work with the SEARCH App revealed that image stabilization would be required for optimum performance.MethodsTo improve image capture, we designed and tested a device (the ‘Donut’), which standardizes distance, angle, rotation and lighting. We then ran an experimental trial with 194 participants to measure the impact of the Donut on identification rates. Images of the participant’s left ear were taken both with and without use of the Donut, then processed by the SEARCH algorithm, measuring the top one and top ten most likely matches.ResultsWith the Donut, the top one identification rate and top ten identification rates were 99.5 and 99.5%, respectively, vs. 38.4 and 24.1%, respectively, without the Donut (P < 0.0001 for each comparison). In sensitivity analyses, crop technique during pre-processing of images had a powerful impact on identification rates, but this too was facilitated through the Donut.ConclusionsBy standardizing lighting, angle and spatial location of the ear, the Donut achieved near perfect identification rates on a cohort of 194 participants, proving the feasibility and effectiveness of using the ear as a biometric identifier.Trial registrationThis study did not include a medical intervention or assess a medical outcome, and therefore did not meet the definition of a human subjects research study as defined by FDAAA. We did not register our study under clinicaltrials.gov.
Highlights
In many low and middle-income countries (LMICs), difficulties in patient identification are a major obstacle to the delivery of longitudinal care
As in many low and middle-income countries (LMICs), she has no formal mailing address; she is too young to have been named; her family is too poor to participate in the national insurance system; and since she was delivered at home, a still-frequent occurrence in many countries, her birth was never formally registered, leaving her without a birth certificate or social security number
Under previous work through Project Scanning Ears for Child Health (SEARCH), a student team from the Computer Science Department at Boston University analyzed three common feature extraction methods and their accuracy when applied to ear biometric identification: Local Binary Patterns (LBPs), Generic Fourier Descriptor (GFD), and Scale Invariant Feature Transform (SIFT)
Summary
In many low and middle-income countries (LMICs), difficulties in patient identification are a major obstacle to the delivery of longitudinal care. As in many low and middle-income countries (LMICs), she has no formal mailing address; she is too young to have been named (in Zambia, naming is often deferred until 6 weeks of age); her family is too poor to participate in the national insurance system; and since she was delivered at home, a still-frequent occurrence in many countries, her birth was never formally registered, leaving her without a birth certificate or social security number Given this reality, a Zambian infant’s well-child care is coordinated using an ‘Under Five Card’, a 4 × 5 inch hard stock paper trifold which remains the mother’s responsibility to retain. In Zambia, as in many LMICs, the lack of robust patient identification renders even routine medical care fragmented, redundant, porous, and inefficient, and is a major barrier to the effective management of preventable and chronic diseases
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
