An objective system for quantitative assessment of television viewing among children (Family Level Assessment of Screen use in the Home-Television): System development study

Authors: KUMAR VADATHYA, ANIL - Rice University; MUSAAD, SALMA - Children'S Nutrition Research Center (CNRC); BELTRAN, ALICIA - Children'S Nutrition Research Center (CNRC); PEREZ, ORIANA - Children'S Nutrition Research Center (CNRC); MEISTER, LEO - Rice University; BARANOWSKI, TOM - Children'S Nutrition Research Center (CNRC); HUGHES, SHERYL - Children'S Nutrition Research Center (CNRC); MENDOZA, JASON - Fred Hutchinson Cancer Research Center; SABHARWAL, ASHUTOSH - Rice University; VEERARAGHAVAN, ASHOK - Rice University; O'CONNOR, TERESIA - Children'S Nutrition Research Center (CNRC)

Submitted to: JMIR Pediatrics and Parenting
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/11/2022
Publication Date: 3/24/2022
Citation: Kumar Vadathya, A., Musaad, S., Beltran, A., Perez, O., Meister, L., Baranowski, T., Hughes, S.O., Mendoza, J.A., Sabharwal, A., Veeraraghavan, A., O'Connor, T. 2022. An objective system for quantitative assessment of television viewing among children (Family Level Assessment of Screen use in the Home-Television): System development study. JMIR Pediatrics and Parenting. 5(1). Article e33569. https://doi.org/10.2196/33569.
DOI: https://doi.org/10.2196/33569

Interpretive Summary: Children's screen use, including TV, mobile devices, computers and gaming systems, have been associated with several health, developmental and psychosocial outcomes. However, to date almost all studies have relied on child- or parent-report of general estimates of the time children engages with such screens. This calls into question the strength and accuracy of the associations found due to measurement errors in self-report estimates. The aim of this project was to develop an objective and automated system for measuring children's TV viewing to help address the existing problems in measuring children's screen use. Our multi-disciplinary team developed the Family Level Assessment of Screen use in the Home (FLASH)-TV system which relies on video images captured in front of a TV and process the images in three sequential steps using machine learning algorithms. Four observational design studies with a total of 21 families were conducted to provide video data that was labeled for ground truth to train each of three steps of FLASH-TV processing: face detection, face recognition and gaze detection. FLASH-TV demonstrated high sensitivity for detecting faces and performed well on face verification when the child's gaze was on the television. The estimates for child's gaze on TV was moderate to good. When sequentially processing the video data through all three steps to measure the duration of screen viewing, the correlation to ground truth was good. FLASH-TV offers the first objective and automated tool for measuring children's TV viewing. With further enhancements, it promises to be a critical research tool to better identify the health, developmental and psychosocial effects of watching TV or other large screens.

Technical Abstract: Television viewing among children is associated with developmental and health outcomes, yet measurement techniques for television viewing are prone to errors, biases, or both. This study aims to develop a system to objectively and passively measure children's television viewing time. The Family Level Assessment of Screen Use in the Home-Television (FLASH-TV) system includes three sequential algorithms applied to video data collected in front of a television screen: face detection, face verification, and gaze estimation. A total of 21 families of diverse race and ethnicity were enrolled in 1 of 4 design studies to train the algorithms and provide proof of concept testing for the integrated FLASH-TV system. Video data were collected from each family in a laboratory mimicking a living room or in the child's home. Staff coded the video data for the target child as the gold standard. The accuracy, sensitivity, specificity, positive predictive value, and negative predictive value were calculated for each algorithm, as compared with the gold standard. Prevalence and biased adjusted k scores and an intraclass correlation using a generalized linear mixed model compared FLASH-TV's estimation of television viewing duration to the gold standard. FLASH-TV demonstrated high sensitivity for detecting faces (95.5%-97.9%) and performed well on face verification when the child’s gaze was on the television. Each of the metrics for estimating the child's gaze on the screen was moderate to good (range: 55.1% negative predictive value to 91.2% specificity). When combining the 3 sequential steps, FLASH-TV estimation of the child's screen viewing was overall good, with an intraclass correlation for an overall time watching television of 0.725 across conditions. FLASH-TV offers a critical step forward in improving the assessment of children's television viewing.