The linkage between the perception of neighbourhood and physical activity in Guangzhou, China: using street view imagery with deep learning techniques

Integrating street view images and deep learning to explore the association between human perceptions of the built environment and cardiovascular disease in older adults

Understanding how built environment attributes affect health remains important. While many studies have explored the objective characteristics of built environments that affect health outcomes, few have examined the role of human perceptions of built environments on physical health. Baidu Street View images and computer vision technological advances have helped researchers overcome the constraints of traditional methods of measuring human perceptions (e.g., these methods are laborious, time-consuming, and costly), allowing for large-scale measurements of human perceptions. This study estimated human perceptions of the built environment (e.g., beauty, boredom, depression, safety, vitality, and wealth) by adopting Baidu Street View images and deep learning algorithms. Negative binomial regression models were employed to analyze the relationship between human perceptions and cardiovascular disease in older adults (e.g., ischemic heart disease and cerebrovascular disease). The results indicated that wealth perception is negatively related to the risk of cardiovascular disease. However, depression and vitality perceptions are positively associated with the risk of cardiovascular disease. Furthermore, we found no relationship between beauty, boredom, safety perceptions, and the risk of cardiovascular disease. Our findings highlight the importance of human perceptions in the development of healthy city planning and facilitate a comprehensive understanding of the relationship between built environment characteristics and health outcomes in older adults. They also demonstrate that street view images have the potential to provide insights into this complicated issue, assisting in the formulation of refined interventions and health policies.

The influence of neighborhood quality on tourism in China: Using Baidu Street View pictures and deep learning techniques

Previous studies have investigated the determinants of urban tourism development from the various attributes of neighborhood quality. However, traditional methods to assess neighborhood quality are often subjective, costly, and only on a small scale. To fill this research gap, this study applies the recent development in big data of street view images, deep learning algorithms, and image processing technology to assess quantitatively four attributes of neighborhood quality, namely street facilities, architectural landscape, green or ecological environment, and scene visibility. The paper collects more than 7.8 million Baidu SVPs of 232 prefecture-level cities in China and applies deep learning techniques to recognize these images. This paper then tries to examine the influence of neighborhood quality on regional tourism development. Empirical results show that both levels of street facilities and greenery environment promote tourism. However, the construction intensity of the landscape has an inhibitory influence on the development of tourism. The threshold test shows that the intensity of the influence varies with the city's overall economic level. These conclusions are of great significance for the development of China's urban construction and tourism economy, and also provide a useful reference for policymakers. The methodological procedure is reduplicative and can be applied to other challenging cases.

A Framework of Community Pedestrian Network Design Based on Urban Network Analysis

Community is the foundation of modern cities, where urban residents spend most of their lifetime. Effective and healthy community design plays a vital role in improving residents’ living quality. Pedestrian network is an indispensable element in the community. Successful pedestrian network design can help the residents be healthy both physically and mentally, build the awareness of “Go Green” for the society, and finally contribute to low-carbon and green cities. This paper proposes a community pedestrian network design method based on Urban Network Analysis with the help of the Rhino software. A case study of a typical community in Guangzhou, China was implemented, specifying the steps of the proposed method. The findings presented include the features of the citizens and the accessibilities of the neighbors that are obtained from the community pedestrian network simulation. The limitation and scalability of this method was discussed. The proposed method can be essential to designing healthy and sustainable communities.

Analysis of developments and hotspots of international research on sports AI

In this paper, 1,538 papers retrieved with the keywords “sports artificial intelligence (AI)” on the Web of Science database since 2007 were taken as the data source, and the Cite Space V software was used to visualize and analyze them. A visual knowledge graph was used to streamline the countries, institutions and authors conducting sports AI research, discipline distribution, research hotspots and development trends in the past 15 years. Subsequently, its development direction and research progress were discussed. Sports AI was widely distributed, with the US, China and the UK leading the way. The most prolific authors and teams in research on sports AI were concentrated in American universities. Their main research direction is to develop and improve smart wearable devices based on machine learning and deep learning technologies for different groups of people. Research on sports AI involved multiple disciplines, which mainly applied and referred to research methodologies and theories on engineering, computer science and sports science. It could be seen from the frequency and centrality of keywords that in the current field of sports AI, machine learning is the main direction, artificial neural networks is the main algorithm, and practical and empirical research based on data mining is the focus. The research hotspots were divided into three major clusters: physical health promotion, sports injury prevention and control, and athletic performance enhancement. How to introduce intelligent technology into sports for a perfect integration still has an arduous and long way to go. Future development requires joint efforts and participation of scientific researchers, professionals and common people.

Understanding the role of urban social and physical environment in opioid overdose events using found geospatial data

Opioid use disorder is a serious public health crisis in the United States. Manifestations such as opioid overdose events (OOEs) vary within and across communities and there is growing evidence that this variation is partially rooted in community-level social and economic conditions. The lack of high spatial resolution, timely data has hampered research into the associations between OOEs and social and physical environments. We explore the use of non-traditional, "found" geospatial data collected for other purposes as indicators of urban social-environmental conditions and their relationships with OOEs at the neighborhood level. We evaluate the use of Google Street View images and non-emergency "311" service requests, along with US Census data as indicators of social and physical conditions in community neighborhoods. We estimate negative binomial regression models with OOE data from first responders in Columbus, Ohio, USA between January 1, 2016, and December 31, 2017. Higher numbers of OOEs were positively associated with service request indicators of neighborhood physical and social disorder and street view imagery rated as boring or depressing based on a pre-trained random forest regression model. Perceived safety, wealth, and liveliness measures from the street view imagery were negatively associated with risk of an OOE. Age group 50-64 was positively associated with risk of an OOE but age 35-49 was negative. White population, percentage of individuals living in poverty, and percentage of vacant housing units were also found significantly positive however, median income and percentage of people with a bachelor's degree or higher were found negative. Our result shows neighborhood social and physical environment characteristics are associated with likelihood of OOEs. Our study adds to the scientific evidence that the opioid epidemic crisis is partially rooted in social inequality, distress and underinvestment. It also shows the previously underutilized data sources hold promise for providing insights into this complex problem to help inform the development of population-level interventions and harm reduction policies.

Effect of Urban Green Space in the Hilly Environment on Physical Activity and Health Outcomes: Mediation Analysis on Multiple Greenery Measures

Background: Green spaces reduce the risk of multiple adverse health outcomes by encouraging physical activity. This study examined correlations between urban green space and residents’ health outcomes in hilly neighborhoods: if they are mediated by social cohesion, visual aesthetics, and safety. Methods: We used multiple green space indicators, including normalized difference vegetation index (NDVI) extracted from satellite imagery, green view index (GVI) obtained from street view data using deep learning methods, park availability, and perceived level of greenery. Hilly terrain was assessed by the standard deviation of the elevation to represent variations in slope. Resident health outcomes were quantified by their psychological and physiological health as well as physical activity. Communities were grouped by quartiles of slopes. Then a mediation model was applied, controlling for socio-demographic factors. Results: Residents who perceived higher quality greenery experienced stronger social cohesion, spent more time on physical activity and had better mental health outcomes. The objective greenery indicators were not always associated with physical activity and might have a negative influence with certain terrain. Conclusions: Perceived green space offers an alternative explanation of the effects on physical activity and mental health in hilly neighborhoods. In some circumstances, geographical environment features should be accounted for to determine the association of green space and resident health outcomes.

Associations Between Street-View Perceptions and Housing Prices: Subjective vs. Objective Measures Using Computer Vision and Machine Learning Techniques

This study investigated the extent to which subjectively and objectively measured street-level perceptions complement or conflict with each other in explaining property value. Street-scene perceptions can be subjectively assessed from self-reported survey questions, or objectively quantified from land use data or pixel ratios of physical features extracted from street-view imagery. Prior studies mainly relied on objective indicators to describe perceptions and found that a better street environment is associated with a price premium. While very few studies have addressed the impact of subjectively-assessed perceptions. We hypothesized that human perceptions have a subtle relationship to physical features that cannot be comprehensively captured with objective indicators. Subjective measures could be more effective to describe human perceptions, thus might explain more housing price variations. To test the hypothesis, we both subjectively and objectively measured six pairwise eye-level perceptions (i.e., Greenness, Walkability, Safety, Imageability, Enclosure, and Complexity). We then investigated their coherence and divergence for each perception respectively. Moreover, we revealed their similar or opposite effects in explaining house prices in Shanghai using the hedonic price model (HPM). Our intention was not to make causal statements. Instead, we set to address the coherent and conflicting effects of the two measures in explaining people’s behaviors and preferences. Our method is high-throughput by extending classical urban design measurement protocols with current artificial intelligence (AI) frameworks for urban-scene understanding. First, we found the percentage increases in housing prices attributable to street-view perceptions were significant for both subjective and objective measures. While subjective scores explained more variance over objective scores. Second, the two measures exhibited opposite signs in explaining house prices for Greenness and Imageability perceptions. Our results indicated that objective measures which simply extract or recombine individual streetscape pixels cannot fully capture human perceptions. For perceptual qualities that were not familiar to the average person (e.g., Imageability), a subjective framework exhibits better performance. Conversely, for perceptions whose connotation are self-evident (e.g., Greenness), objective measures could outperform the subjective counterparts. This study demonstrates a more holistic understanding for street-scene perceptions and their relations to property values. It also sheds light on future studies where the coherence and divergence of the two measures could be further stressed.

Measuring the association of objective and perceived neighborhood environment with physical activity in older adults: challenges and implications from a systematic review

BACKGROUND

A supportive environment is a key factor in addressing the issue of health among older adults. There is already sufficient evidence that objective and self-reported measures of the neighborhood environment should be taken into account as crucial components of active aging, as they have been shown to influence physical activity; particularly in people aged 60+. Thus, both could inform policies and practices that promote successful aging in place. An increasing number of studies meanwhile consider these exposures in analyzing their impact on physical activity in the elderly. However, there is a wide variety of definitions, measurements and methodological approaches, which complicates the process of obtaining comparable estimates of the effects and pooled results. The aim of this review was to identify and summarize these differences in order to emphasize methodological implications for future reviews and meta analyzes in this field and, thus, to create a sound basis for synthesized evidence.

METHODS

A systematic literature search across eight databases was conducted to identify peer-reviewed articles examining the association of objective and perceived measures of the neighborhood environment and objectively measured or self-reported physical activity in adults aged ≥ 60 years. Two authors independently screened the articles according to predefined eligibility criteria, extracted data, and assessed study quality. A qualitative synthesis of the findings is provided.

RESULTS

Of the 2967 records retrieved, 35 studies met the inclusion criteria. Five categories of methodological approaches, numerous measurement instruments to assess the neighborhood environment and physical activity, as well as several clusters of definitions of neighborhood, were identified.

CONCLUSIONS

The strength of evidence of the associations of specific categories of environmental attributes with physical activity varies across measurement types of the outcome and exposures as well as the physical activity domain observed and the operationalization of neighborhood. The latter being of great importance for the targeted age group. In the light of this, future reviews should consider these variations and stratify their summaries according to the different approaches, measures and definitions. Further, underlying mechanisms should be explored.

Objective scoring of streetscape walkability related to leisure walking: Statistical modeling approach with semantic segmentation of Google Street View images

Although the pedestrian-friendly qualities of streetscapes promote walking, quantitative understanding of streetscape functionality remains insufficient. This study proposed a novel automated method to assess streetscape walkability (SW) using semantic segmentation and statistical modeling on Google Street View images. Using compositions of segmented streetscape elements, such as buildings and street trees, a regression-style model was built to predict SW, scored using a human-based auditing method. Older female active leisure walkers living in Bunkyo Ward, Tokyo, are associated with SW scores estimated by the model (OR = 3.783; 95% CI = 1.459 to 10.409), but male walkers are not.

Viewing an alpine environment positively affects emotional analytics in patients with somatoform, depressive and anxiety disorders as well as in healthy controls

BACKGROUND

Patients with somatoform, depressive or anxiety disorders often don't respond well to medical treatment and experience many side effects. It is thus of clinical relevance to identify alternative, scientifically based, treatments. Our approach is based on the recent evidence that urbanicity has been shown to be associated with an increased risk for mental disorders. Conversely, green and blue environments show a dose-dependent beneficial impact on mental health.

METHODS

Here we evaluate the effect of viewing stimuli of individuals in an alpine environment on emotional analytics in 183 patients with psychiatric disorders (mostly somatoform, depressive and anxiety disorders) and 315 healthy controls (HC). Emotional analytics (valence: unhappy vs happy, arousal: calm vs excited, dominance: controlled vs in control) were assessed using the Self-Assessment Manikin. Further parameters related to mental health and physical activity were recorded.

RESULTS

Emotional analytics of patients indicated that they felt less happy, less in control and had higher levels of arousal than HC when viewing neutral stimuli. The comparison alpine>neutral stimuli showed a significant positive effect of alpine stimuli on emotional analytics in both groups. Patients and HC both felt attracted to the scenes displayed in the alpine stimuli. Emotional analytics correlated positively with resilience and inversely with perceived stress.

CONCLUSIONS

Preventive and therapeutic programs for patients with somatoform, depressive and anxiety disorders should consider taking the benefits of natural outdoor environments, such as alpine environments, into account. Organizational barriers which are preventing the implementation of such programs in clinical practice need to be identified and addressed.

The association between spatial access to physical activity facilities within home and workplace neighborhoods and time spent on physical activities: evidence from Guangzhou, China

BACKGROUND

Urban residents from the developing world have increasingly adopted a sedentary lifestyle and spend less time on physical activities (PA). Previous studies on the association between PA facilities and individuals' PA levels are based on the assumption that individuals have opportunities to use PA facilities within neighborhoods all day long, ignoring the fact that their willingness and opportunities to use nearby facilities depend on how much discretionary time (any time when people have a choice what to do) they have. Further, scant attention has been paid to the influence of PA facilities within both residential and workplace neighborhoods in the dense urban context. To address the above research gaps, this study investigated the links between the spatial access to PA facilities within home/workplace neighborhoods and time spent on PA among working adults, focusing on whether results were different when different measures of accessibility were used and whether participants' discretionary time over a week affected their time spent on PA.

METHOD

This study used data from a questionnaire survey (n = 1002) in Guangzhou between June and July 2017 and point of interest (POI) data from online mapping resources. Outcome variables included the amount of time spent on physical activity/moderate and vigorous intensity physical activity (PA/MVPA) over the past week. Home/workplace neighborhoods were measured as different distance buffers (500 m circular buffers, 1000 m circular buffers, and 1080 m network buffers) around each respondent's home/workplace. Spatial access to PA facilities was measured using two indicators: the counts of PA facilities and proximity to PA facilities within home/workplace neighborhoods. The amount of discretionary time was calculated based on activity log data of working day/weekend day from the Guangzhou questionnaire survey, and regression models were used to examine relationships between the spatial access of PA facilities, the time spent on PA/MVPA, and the amount of discretionary time, adjusted for covariates. Associations were stratified by gender, age, education, and income.

RESULTS

Using different measures of accessibility (the counts of and proximity to PA facilities) generated different results. Specifically, participants spent more time on PA/MVPA when they lived in neighborhoods with more PA facilities and spent more time on MVPA when worked in closer proximity to PA facilities. A larger amount of discretionary time was associated with more time spent on PA/MVPA, but it did not strengthen the relationship between access to PA facilities and PA/MVPA time. In addition, relationships between access to PA facilities and PA levels varied by gender, age, education, and income.

CONCLUSION

This study contributes to the knowledge of PA-promoting environments by considering both the home and workplace contexts and by taking into account the temporal attributes of contextual influences. Policymakers and urban planners are advised to take into account the workplace context and the temporal variability of neighborhood influences when allocating public PA facilities and public spaces.

Spatial predictive properties of built environment characteristics assessed by drop-and-spin virtual neighborhood auditing

Background

Virtual neighborhood audits have been used to visually assess characteristics of the built environment for health research. Few studies have investigated spatial predictive properties of audit item responses patterns, which are important for sampling efficiency and audit item selection. We investigated the spatial properties, with a focus on predictive accuracy, of 31 individual audit items related to built environment in a major Metropolitan region of the Northeast United States.

Methods

Approximately 8000 Google Street View (GSV) scenes were assessed using the CANVAS virtual audit tool. Eleven trained raters audited the 360^° view of each GSV scene for 10 sidewalk-, 10 intersection-, and 11 neighborhood physical disorder-related characteristics. Nested semivariograms and regression Kriging were used to investigate the presence and influence of both large- and small-spatial scale relationships as well as the role of rater variability on audit item spatial properties (measurement error, spatial autocorrelation, prediction accuracy). Receiver Operator Curve (ROC) Area Under the Curve (AUC) based on cross-validated spatial models summarized overall predictive accuracy. Correlations between predicted audit item responses and select demographic, economic, and housing characteristics were investigated.

Results

Prediction accuracy was better within spatial models of all items accounting for both small-scale and large- spatial scale variation (vs large-scale only), and further improved with additional adjustment for rater in a majority of modeled items. Spatial predictive accuracy was considered ‘Excellent’ (0.8 ≤ ROC AUC < 0.9) for full models of all but four items. Predictive accuracy was highest and improved the most with rater adjustment for neighborhood physical disorder-related items. The largest gains in predictive accuracy comparing large- + small-scale to large-scale only models were among intersection- and sidewalk-items. Predicted responses to neighborhood physical disorder-related items correlated strongly with one another and were also strongly correlated with racial-ethnic composition, socioeconomic indicators, and residential mobility.

Conclusions

Audits of sidewalk and intersection characteristics exhibit pronounced variability, requiring more spatially dense samples than neighborhood physical disorder audits do for equivalent accuracy. Incorporating rater effects into spatial models improves predictive accuracy especially among neighborhood physical disorder-related items.