Biometric recognition of newborns and young children for vaccinations and health care: a non-randomized prospective clinical trial.

Although universal biometrics have been broadly called for, and there are many validated technologies to recognize adults, these technologies have been ineffective in newborns and young children. The present work describes the development and clinical testing of a fingerprint capture system for longitudinal biometric recognition of newborns and young children to support vaccination and clinical follow-up. The reader consists of a high-resolution monochromatic imaging system with an ergonomic industrial design to comfortably support and align infant fingers for imaging without a platen. This imaging approach without a platen, also called free-space imaging, reduces fingerprint distortion and ensures a more consistent finger placement. This system was tested in a newborn ward and immunization clinic at an urban hospital in Baja, California, Mexico, from 2017 to 2019. Nearly five hundred children were enrolled and followed for up to 24 months. With a protocol of imaging all ten fingers, the failure to enroll (FTE) rate was < 1% when acquiring at least two fingers for all ages and < 2% when enrolling at least four fingers. The verification (1:1) true accept rate (TAR) was 77% for newborns enrolled at ≤ 3 days of age and 96% for those enrolled at ≥ 4 days of age, both at a time gap of 15-30 days after enrollment at a false accept rate (FAR) of 0.1%. Using the top-ranked match score, the identification rate (1:many) was 86% for the ≤ 3 days enrollment age and 97% for age ≥ 4 days for a single finger at 15-30 days after enrollment. The enrollment protocol and the frequency of updating will increase for infants compared to adults. However, these data suggest that a high-resolution, free space imaging technique may fill the final gap for universal biometrics across all populations called for by the United Nations Sustainable Development Goal 16.9.

Biometric recognition of newborns and infants by non-contact fingerprinting: lessons learned.

Despite years of effort, reliable biometric identification of newborns and young children has remained elusive. In this paper, we review the importance of trusted identification methods, the biometric landscape for infants and adults, barriers and success stories, and we discuss specific failure modes particular to young children. We then describe our approach to infant identification using non-contact optical imaging of fingerprints. We detail our technology development history, including Human-Centered Design methods, various iterations of our platform, and how these iterations addressed failure modes in the identification process. We close with a brief description of our clinical trial of newborns and infants at an urban hospital in Mexico and report preliminary results that show high accuracy, with matching rates consistent with acceptable field-performance for reliable biometric identification in large populations.

Initial neural net construction for the detection of cervical intraepithelial neoplasia by fluorescence imaging.

OBJECTIVE: The aim of this study was to initiate neural net construction for the detection of cervical intraepithelial neoplasia by fluorescence imaging. STUDY DESIGN: Thirty-three women with abnormal Papanicolaou smears underwent fluorescence imaging during colposcopy. With the use of >4000 training pixels and >1000 test pixels, intrapatient nets were constructed from the spectral data of 17 women. An interpatient net that discriminated between cervical intraepithelial neoplasia 1 and normal tissue classes among patients was constructed with the use of >2300 training pixels and >2000 test pixels from 12 women. Average correct classification rates were determined. Sensitivities, specificities, and positive and negative predictive values for cervical intraepithelial neoplasia grade 1 and normal tissue classes were calculated. Extrapolated false-color cervical images were generated. RESULTS: Average correct classification rates were 96.5% for the intrapatient nets and 97.5% for the interpatient net. The sensitivity, specificity, and positive and negative predictive values for cervical intraepithelial neoplasia grade 1 were 98.2%, 98.9%, 71.4%, and 99.9%, respectively. CONCLUSION: Initial results suggest that neural nets that are constructed from fluorescence imaging spectra may offer a potential method for the detection of cervical intraepithelial neoplasia.