+1 Recommend
0 collections
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

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


      Read this article at

          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.



          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.


          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.


          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.


          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.


          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.


          • Deep learning and street view images were used to assess streetscape green and blue space.

          • Street view green and blue spaces were uncorrelated with satellite-derived metrics (i.e., NDVI, NDWI, and GlobeLand30).

          • The mental health of elderly people seemed enhanced by exposure to street view green and blue spaces.

          • No evidence of depression-green and blue space associations when remote sensing-based metrics were used.

          • People-centric exposure assessments using street view data provide great potential for environmental health studies.

          Related collections

          Most cited references57

          • Record: found
          • Abstract: found
          • Article: not found

          Green space, urbanity, and health: how strong is the relation?

          To investigate the strength of the relation between the amount of green space in people's living environment and their perceived general health. This relation is analysed for different age and socioeconomic groups. Furthermore, it is analysed separately for urban and more rural areas, because the strength of the relation was expected to vary with urbanity. The study includes 250 782 people registered with 104 general practices who filled in a self administered form on sociodemographic background and perceived general health. The percentage of green space (urban green space, agricultural space, natural green space) within a one kilometre and three kilometre radius around the postal code coordinates was calculated for each household. Multilevel logistic regression analyses were performed at three levels-that is, individual level, family level, and practice level-controlled for sociodemographic characteristics. The percentage of green space inside a one kilometre and a three kilometre radius had a significant relation to perceived general health. The relation was generally present at all degrees of urbanity. The overall relation is somewhat stronger for lower socioeconomic groups. Elderly, youth, and secondary educated people in large cities seem to benefit more from presence of green areas in their living environment than other groups in large cities. This research shows that the percentage of green space in people's living environment has a positive association with the perceived general health of residents. Green space seems to be more than just a luxury and consequently the development of green space should be allocated a more central position in spatial planning policy.
            • Record: found
            • Abstract: not found
            • Article: not found

            Global land cover mapping at 30m resolution: A POK-based operational approach

              • Record: found
              • Abstract: found
              • Article: found
              Is Open Access

              Exposure to Neighborhood Green Space and Mental Health: Evidence from the Survey of the Health of Wisconsin

              Green space is now widely viewed as a health-promoting characteristic of residential environments, and has been linked to mental health benefits such as recovery from mental fatigue and reduced stress, particularly through experimental work in environmental psychology. Few population level studies have examined the relationships between green space and mental health. Further, few studies have considered the role of green space in non-urban settings. This study contributes a population-level perspective from the United States to examine the relationship between environmental green space and mental health outcomes in a study area that includes a spectrum of urban to rural environments. Multivariate survey regression analyses examine the association between green space and mental health using the unique, population-based Survey of the Health of Wisconsin database. Analyses were adjusted for length of residence in the neighborhood to reduce the impact of neighborhood selection bias. Higher levels of neighborhood green space were associated with significantly lower levels of symptomology for depression, anxiety and stress, after controlling for a wide range of confounding factors. Results suggest that “greening” could be a potential population mental health improvement strategy in the United States.

                Author and article information

                Environ Int
                Environ Int
                Environment International
                Elsevier Science
                1 May 2019
                May 2019
                : 126
                : 107-117
                [a ]Department of Human Geography and Spatial Planning, Utrecht University, Utrecht, The Netherlands
                [b ]School of Information Engineering, China University of Geosciences, Wuhan, China
                [c ]School of Geography and Planning, Sun Yat-Sen University, Guangzhou, China
                [d ]Guangdong Key Laboratory for Urbanization and Geo-Simulation, Sun Yat-Sen University, Guangzhou, China
                Author notes
                [* ]Corresponding author at: School of Information Engineering, China University of Geosciences, Wuhan, 430074, China; School of Geography and Planning, Sun Yat-Sen University, Xingang Xi Road, Guangzhou 510275, China. wangry6@ 123456mail2.sysu.edu.cn

                These authors contributed equally.

                © 2019 The Authors

                This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

                : 27 October 2018
                : 31 January 2019
                : 3 February 2019

                deep learning,street view data,natural environments,exposures,depression,the elderly,china


                Comment on this article