![]() ![]() In order to develop an effective infant fingerprint platform, it is critical to test in realistic environments on infants of all ages. This report highlights key issues we encountered, and the steps taken to resolve them to achieve a robust infant biometric system. Through this process, we accumulated numerous lessons-learned, resulting in optimized procedures and a system design that enabled enrollment and scanning even in demanding field conditions. We worked with children from birth through 18 months, along with parents, caregivers, nurses, doctors, health officials and vaccinators, to observe infant behavior and caregiver interaction through many stages of iteration of both contact and non-contact system designs. We employed Human-centered Design principles of early stakeholder engagement and co-design, problem reframing, and agile prototyping, to rethink the infant ID problem 12– 14. Here we report our design strategy, detailed analysis of failure modes for infant fingerprinting, and key insights and requirements that can instruct development of reliable and usable infant-centric biometrics. ![]() This report is meant to accompany subsequent publications detailing the performance of our non-contact technology in clinical testing. Based on testing in laboratory and clinical settings we concluded that a non-contact imaging method was the best for newborns and infants. Using this method, we compared two contact-based approaches, frustrated total internal reflection (FTIR), non-FTIR direct imaging, and non-contact imaging approach with multiple interaction designs 11. To test this hypothesis, and potentially develop a reliable infant biometric, we developed a modular biometric prototyping platform that provided a common imaging back-end to be coupled with various front-ends allowing a wide variety of infant-device-practitioner interactions. We hypothesized that the malleability of infant skin coupled with grasping and other infant reflexes were leading to deformation of the fingerprint by current biometric scanner platens themselves, which explains why even higher resolution platforms have still failed with children under six months of age. To date none of these has shown to be reliable for newborn and very young infant enrollment, and only fingerprinting has shown promise in terms of universality, acceptability, persistence over time from birth 5 and interoperability across acquisition methods 6– 10. These included eye scanning, palm vein scanning, ear and face recognition, and finger and palm-based methods. We investigated an array of biometric methods for infant identification. There have been numerous attempts to utilize standard fingerprint technologies with infants with limited success and while new technologies have been developed to address the problem, improvements have been limited to children over 6 months of age 4. Today one of the primary barriers to fulfilling SDG 16.9 has been the lack of universal biometric technology able to reliably identify newborns, young children and even at times adults 3. To address the need, United Nations Sustainable Development Goal 16.9 calls to provide legal identity for all, including free birth registrations by 2030 2. Globally over 1 billion people lack legal identification and almost half of them are infants and children 1. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |