Visible to the public S&CC: Promoting a Healthier Urban Community: Prioritization of Risk Factors for the Prevention and Treatment of Pediatric ObesityConflict Detection Enabled

Project Details
Lead PI:Ming Dong
Co-PI(s):Dongxiao Zhu
Elizabeth Kuhl
Performance Period:09/01/16 - 08/31/18
Institution(s):Wayne State University
Sponsor(s):National Science Foundation
Award Number:1637312
390 Reads. Placed 458 out of 803 NSF CPS Projects based on total reads on all related artifacts.
Abstract: Urban communities are facing many challenges due to the increasing complexity of urban life, declining urban services and growing health and economic disparities. While diverse stakeholders are engaged in understanding and solving these issues, progress has not been commensurate with the effort, attributed partially to the limited collaboration and data sharing. The persistence of obesity disparities in early childhood is one example of the negative consequences of such isolated efforts. Obesity is a multi-faced health outcome. While some risk-factors for obesity are universal, others are highly specific to the community in which a particular child lives. As such, successful efforts to prevent and treat pediatric obesity depend upon integration of data from multiple community sources and systems. The overall objective of the proposed research is to develop an innovative data-driven health informatics system (Preschool Risk for Obesity Portal; PROP) that aims to promote comprehensive, efficient, and personalized obesity-related care for preschoolers living in urban communities. Through the data sharing and integration within the community and the development, along with the beta-test of PROP, the project has the potential to promote a healthier urban community. Through the data sharing and integration within the community and the development, along with the beta-test of PROP, the project has the potential to promote a healthier urban community. The approach taken could be adapted for older pediatric age-groups, adults, and to address other health disparity issues in urban communities. From a technical perspective, the PIs will: 1) design innovative multi-level mixed effects machine learning methods and scalable algorithms that can precisely identify and prioritize a preschooler's personalized risk factors for obesity and 2) develop a data- and tool-rich online system dedicated to pediatric obesity. Specifically, design (Phase one) and proof-of-concept testing (Phase two) for the PROP algorithms will be completed in this exploratory work. After the successful completion, the second component of PROP (an eHealth intervention) will be developed in a separate, bigger project for efficacy trial (Phase three) and effectiveness research (Phase four). The significant intellectual merit of this project lies in the novel algorithms for information extraction and understanding from multi-scale, correlated, and heterogeneous datasets. The online system dedicated to pediatric obesity will be built for the rapid dissemination of core computational techniques to researchers.