Deep Learning to Unveil Correlations between Urban Landscape and Population Health

Abstract

The global healthcare landscape is continuously changing throughout the world as technology advances, leading to a gradual change in lifestyle. Several diseases such as asthma and cardiovascular conditions are becoming more diffuse, due to a rise in pollution exposure and a more sedentary lifestyle. Healthcare providers deal with increasing new challenges, and thanks to fast-developing big data technologies, they can be faced with systems that provide direct support to citizens. In this context, within the EU-funded Participatory Urban Living for Sustainable Environments (PULSE) project, we are implementing a data analytic platform designed to provide public health decision makers with advanced approaches, to jointly analyze maps and geospatial information with healthcare and air pollution data. In this paper we describe a component of such platforms, which couples deep learning analysis of urban geospatial images with healthcare indexes collected by the 500 Cities project. By applying a pre-learned deep Neural Network architecture, satellite images of New York City are analyzed and latent feature variables are extracted. These features are used to derive clusters, which are correlated with healthcare indicators by means of a multivariate classification model. Thanks to this pipeline, it is possible to show that, in New York City, health care indexes are significantly correlated to the urban landscape. This pipeline can serve as a basis to ease urban planning, since the same interventions can be organized on similar areas, even if geographically distant.

Impact of COVID-19 lockdown on PM concentrations in an Italian Northern City: A year-by-year assessment

“…One of the key innovation aspects of PULSE is the use of data with high spatial and temporal resolution. The increase in terms of resolution can be beneficial to study phenomena of public health interest at a neighborhood level, thus taking into account social and health disparities that often characterize large urban environments, as highlighted by studies we performed in the context of the project [ 13 , 14 ]. Despite this necessity, urban data is rarely collected with a sufficient spatial granularity, due to the high costs and difficulties of the process.…”

The Adoption of Digital Technologies and Artificial Intelligence in Urban Health: A Scoping Review

“…Europe was the most represented continent, with six studies, followed by Asia and North America (Figure 2). Among those included, only one article out of twelve (Pala et al) [25] investigated both the use of AI and digital technologies to enhance the exposure conditions to health determinants, through the analysis of prevention, clinical, and behavioral outcomes, and the related impact on population health, whereas the other eleven studies, out of twelve, focused on the adoption of AI and digital technologies to improve only the exposure con- Among those included, only one article out of twelve (Pala et al) [25] investigated both the use of AI and digital technologies to enhance the exposure conditions to health determinants, through the analysis of prevention, clinical, and behavioral outcomes, and the related impact on population health, whereas the other eleven studies, out of twelve, focused on the adoption of AI and digital technologies to improve only the exposure conditions to health determinants, without putting direct attention to population health impact. The vast majority (75%) of the included articles were published in highly specific academic journals regarding science and technology of sensors and their applications.…”

“…The paper by Pala et al [25] demonstrated that a data analytic platform, based on a pre-learned deep Neural Network architecture, can provide policymakers with advanced approaches to analyze maps and geospatial information with healthcare and air pollution data and, therefore, can ease the urban planning process.…”

The Key Role of Geographic Information in Exposomics: The Example of the H2020 Pulse Project

“…The WegGIS and the GWR demonstrate how the use of geographic information can play a fundamental role in representing data and analyzing it in order to create predictive models, in addition, spatial enablement in PULSE is used also as an active tool for the optimization of the intervention strategy. A first example of this, is another study conducted in PULSE and published in April 2020 (Pala et al 2020) on the journal Sensors, were we used a deep learning approach to analyze the relation between urban structure and several public health indexes. This study proposes an analysis pipeline to optimize the intervention design strategies following the idea that if there is a correlation between urban landscape and health indexes, then areas with similar urban structure will have a similar health situation and respond in a similar way to the same kind of interventions.…”