A picture tells a thousand…exposures: Opportunities and challenges of deep learning image analyses in exposure science and environmental epidemiology

Classifying Street Spaces with Street View Images for a Spatial Indicator of Urban Functions

Streets, as one type of land use, are generally treated as developed or impervious areas in most of the land-use/land-cover studies. This coarse classification substantially understates the value of streets as a type of public space with the most complexity. Street space, being an important arena for urban vitality, is valued by various dimensions, such as transportation, recreation, aesthetics, public health, and social interactions. Traditional remote sensing approaches taking a sky viewpoint cannot capture these dimensions not only due to the resolution issue but also the lack of a citizen viewpoint. The proliferation of street view images provides an unprecedented opportunity to characterize street spaces from a citizen perspective at the human scale for an entire city. This paper aims to characterize and classify street spaces based on features extracted from street view images by a deep learning model of computer vision. A rule-based clustering method is devised to support the empirically generated classification of street spaces. The proposed classification scheme of street spaces can serve as an indirect indicator of place-related functions if not a direct one, once its relationship with urban functions is empirically tested and established. This approach is empirically applied to Beijing city to demonstrate its validity.

A population health perspective on artificial intelligence

The burgeoning field of Artificial Intelligence (AI) has the potential to profoundly impact the public’s health. Yet, to make the most of this opportunity, decision-makers must understand AI concepts. In this article, we describe approaches and fields within AI and illustrate through examples how they can contribute to informed decisions, with a focus on population health applications. We first introduce core concepts needed to understand modern uses of AI and then describe its sub-fields. Finally, we examine four sub-fields of AI most relevant to population health along with examples of available tools and frameworks. Artificial intelligence is a broad and complex field, but the tools that enable the use of AI techniques are becoming more accessible, less expensive, and easier to use than ever before. Applications of AI have the potential to assist clinicians, health system managers, policy-makers, and public health practitioners in making more precise, and potentially more effective, decisions.

Using deep learning to examine street view green and blue spaces and their associations with geriatric depression in Beijing, China

BACKGROUND

Residential green and blue spaces may be therapeutic for the mental health. However, solid evidence on the linkage between exposure to green and blue spaces and mental health among the elderly in non-Western countries is scarce and limited to exposure metrics based on remote sensing images (i.e., land cover and vegetation indices). Such overhead-view measures may fail to capture how people perceive the environment on the site.

OBJECTIVE

This study aimed to compare streetscape metrics derived from street view images with satellite-derived ones for the assessment of green and blue space; and to examine associations between exposure to green and blue spaces as well as geriatric depression in Beijing, China.

METHODS

Questionnaire data on 1190 participants aged 60 or above were analyzed cross-sectionally. Depressive symptoms were assessed through the shortened Geriatric Depression Scale (GDS-15). Streetscape green and blue spaces were extracted from Tencent Street View data by a fully convolutional neural network. Indicators derived from street view images were compared with a satellite-based normalized difference vegetation index (NDVI), a normalized difference water index (NDWI), and those derived from GlobeLand30 land cover data on a neighborhood level. Multilevel regressions with neighborhood-level random effects were fitted to assess correlations between GDS-15 scores and these green and blue spaces exposure metrics.

RESULTS

The average cumulative GDS-15 score was 3.4 (i.e., no depressive symptoms). Metrics of green and blue space derived from street view images were not correlated with satellite-based ones. While NDVI was highly correlated with GlobeLand30 green space, NDWI was moderately correlated with GlobeLand30 blue space. Multilevel regressions showed that both street view green and blue spaces were inversely associated with GDS-15 scores and achieved the highest model goodness-of-fit. No significant associations were found with NDVI, NDWI, and GlobeLand30 green and blue space. Our results passed robustness tests.

CONCLUSION

Our findings provide support that street view green and blue spaces are protective against depression for the elderly in China, yet longitudinal confirmation to infer causality is necessary. Street view and satellite-derived green and blue space measures represent different aspects of natural environments. Both street view data and deep learning are valuable tools for automated environmental exposure assessments for health-related studies.

Measuring social, environmental and health inequalities using deep learning and street imagery

Cities are home to an increasing majority of the world’s population. Currently, it is difficult to track social, economic, environmental and health outcomes in cities with high spatial and temporal resolution, needed to evaluate policies regarding urban inequalities. We applied a deep learning approach to street images for measuring spatial distributions of income, education, unemployment, housing, living environment, health and crime. Our model predicts different outcomes directly from raw images without extracting intermediate user-defined features. To evaluate the performance of the approach, we first trained neural networks on a subset of images from London using ground truth data at high spatial resolution from official statistics. We then compared how trained networks separated the best-off from worst-off deciles for different outcomes in images not used in training. The best performance was achieved for quality of the living environment and mean income. Allocation was least successful for crime and self-reported health (but not objectively measured health). We also evaluated how networks trained in London predict outcomes three other major cities in the UK: Birmingham, Manchester, and Leeds. The transferability analysis showed that networks trained in London, fine-tuned with only 1% of images in other cities, achieved performances similar to ones from trained on data from target cities themselves. Our findings demonstrate that street imagery has the potential complement traditional survey-based and administrative data sources for high-resolution urban surveillance to measure inequalities and monitor the impacts of policies that aim to address them.

A Mobile Air Pollution Monitoring Data Set

Air pollution was observed in Hamilton, Ontario, Canada using monitors installed in a mobile platform from November 2005 up to November 2016. The dataset is an aggregation of several project specific monitoring days, which attempted to quantify air pollution spatial variation under varying conditions or in specific regions. Pollutants observed included carbon monoxide, nitric oxide, nitrogen dioxide, total nitrogen oxides, ground-level ozone, particulate matter concentrations for size cuts of 10 µm, 2.5 µm and 1 µm, and sulfur dioxide. Observations were collected over 114 days, which occurred in varying seasons and months. During sampling, the mobile platform travelled at an average speed of 27 km/h. The samples were collected as one-minute integrated samples and are prepared as line-segments, which include an offset for instrument response time. Sampling occurred on major freeways, highways, arterial and residential roads. This dataset is shared in hopes of supporting research on how to best utilize air pollution observations obtained with mobile air pollution platforms, which is a growing technique in the field of urban air pollution monitoring. We conclude with limitations in the data capture technique and recommendations for future mobile monitoring studies.