Machine Learning Approaches for Measuring Neighborhood Environments in Epidemiologic Studies

Abstract

Purpose of review Innovations in information technology, initiatives by local governments to share administrative data, and growing inventories of data available from commercial data aggregators have immensely expanded the information available to describe neighborhood environments, supporting an approach to research we call Urban Health Informatics. This review evaluates the application of machine learning to this new wealth of data for studies of the effects of neighborhood environments on health. Recent findings Prominent machine learning applications in this field include automated image analysis of archived imagery such as Google Street View images, variable selection methods to identify neighborhood environment factors that predict health outcomes from large pools of exposure variables, and spatial interpolation methods to estimate neighborhood conditions across large geographic areas. Summary In each domain, we highlight successes and cautions in the application of machine learning, particularly highlighting legal issues in applying machine learning approaches to Google’s geo-spatial data.

A Parcel Level Housing Conditions Survey in Academic and Civil Context

“…For instance, the paper “Validity of an Ecometric Neighborhood Physical Disorder Measure Constructed by Virtual Street Audit” (Mooney et al, 2014) introduced a new surveying technique. The survey instrument introduced in this paper was followed by several methodology papers interrogating and expanding the technique (Bader et al, 2015; Mooney, Bader, et al, 2017; Quinn et al, 2016; Rundle et al, 2022), and several papers applying the technique connecting public health issues with the built environment (Joshi et al, 2017; Mooney et al, 2016; Mooney, Joshi et al, 2017). This pattern is not a problem in itself.…”

“…Their rating instrument was designed to incorporate as many different existing survey elements as possible, including elements of the Irvine-Minnesota Inventory, the pedestrian Environment Data Scan, the Maryland Inventory of Design Qualities as well as aspects of the Project on Human Development in Chicago Neighborhoods and the New York Housing and Vacancy Survey (Bader et al, 2015, p. 168). One hundred eighty-seven survey items from 300 dispersed census tracts gathered from photographs led to the Rundle et al (2022) review of various machine learning approaches to “Urban Health Informatics.” An algorithm is trained to perform the visual assessment automatically, and in the second stage, regression techniques are used to data mine the results (Random forest, LASSO, etc.). Promising as this attempt at survey convergence may seem, this research agenda may be a dead end.…”

Accessing Eye-level Greenness Visibility from Open-Source Street View Images: A methodological development and implementation in multi-city and multi-country contexts

“…In particular, GSV imposes restrictions on the use of imagery, including limitations on data analysis and extraction (Google, 2018b(Google, , 2020. As stated by Rundle et al (2022), the restriction on certain uses of Google Maps is not grounded in copyright law, which would potentially allow researchers to invoke the "Fair" use principles. Rather, these limitations are based on the contractual agreement that users must adhere to when accessing Google Maps (Google, 2018a).…”

“…The enforceability of such contracts is currently subject to ongoing legal disputes, making it uncertain. Consequently, researchers engaging in this type of research and the journals publishing the resulting papers assume some legal risk as long as the legal status remains unsettled (Rundle et al, 2022;Stringam et al, 2023). In addition, using such data might hinder the adoption of FAIR (Findable, Accessible, Interoperable, and Reusable) data principles in scienti c research (Wilkinson et al, 2016).…”