Health Geography | Shaoqing Dai

Sensing healthy city: from multimodal geographic data, GeoAI to causal inference

Sun, 19 Jul 2026 09:10:00 +0000

Sensing healthy city: from multimodal geographic data, GeoAI to causal inference

Fri, 17 Jul 2026 09:20:00 +0000

Photos

Presenting at the 10th Symposium on Applications of Geographic Information and Spatial Analysis Technologies in Public Health, East China Normal University, Shanghai, July 2026.

Identification of two-dimensional copper signatures in human blood for bladder cancer with machine learning

Tue, 11 Jan 2022 00:00:00 +0000

Figure 1. Cu concentration and the δ65Cu value in plasma and RBC of BCa patients grouped by cancer grade, cancer stage, age, and gender. Each symbol presents an individual subject. (A and B) Cu concentration in plasma and RBC of BCa patients for different grades. “Low” refers to low-grade BCa (n = 16) and “high” refers to high-grade BCa (n = 21). PA = 0.7910 and PB = 0.8324, Mann Whitney test. (C and D) δ65Cu value in plasma and RBC of BCa patients for different grades. PC = 0.0227 and PD = 0.1854, unpaired Student’s two-tailed t-test. (E and F) Cu concentration in plasma and RBC of BCa patients for different cancer stages (n = 31 for Ta/T1 and n = 4 for T2/T3). PE = 0.9692 and PF = 0.7699, unpaired Student’s two-tailed t-test. (G and H) δ65Cu value in plasma and RBC of BCa patients for different cancer stages. PG = 0.2094 and PH = 0.5604, unpaired Student’s two-tailed t-test. (I and J) Variation of the Cu concentration in plasma and RBC of all subjects with age. (K and L) Variation of the δ65Cu value in plasma and RBC of BCa patients with age. (M and N) Variation of the Cu concentration in plasma and RBC of BCa patients with gender (n = 28 for male and n = 13 for female). PM = 0.1083, unpaired Student’s two-tailed t-test; PN = 0.2924, Mann Whitney test. (O and P) Variation of the δ65Cu value in plasma and RBC of BCa patients with gender. PO = 0.8701 and PP = 0.7349, unpaired Student’s two-tailed t-test.

The direct and interactive impacts of hydrological factors on bacillary dysentery across different geographical regions in central China

Tue, 29 Dec 2020 00:00:00 +0000

Figure 1. The conceptual diagram illustrating the relationship between bacillary dysentery and hydrological factors.

Urban Visual Intelligence

Wed, 01 Jan 2020 00:00:00 +0000

Street view imagery (SVI) provides an eye-level, human-centric record of urban environments, capturing pedestrian-experience features such as greenery, facades, and safety-related cues that satellite imagery cannot observe. This project develops urban visual intelligence: turning massive street-level imagery into measurable, explainable, and actionable urban knowledge. The research covers the full chain from data coverage and generation, visual auditing of built environments, eye-level greenery and multi-source urban analysis, to visual-semantic urban governance with large language models. It is also the methodological backbone of my PhD research on improving obesogenic environmental assessments with advanced geospatial methods.

1 Street view data coverage and generation

The global coverage of SVI is highly uneven: published studies concentrate in the United States and Europe, while developing regions across Asia, Africa, and South America remain systematically underrepresented, which limits the inclusiveness of SVI-based analytics. We proposed GeoIdentity-Sat2Street, a geographic identity preserving framework that generates street view imagery from satellite views by coupling a polar-transformation conditional GAN with a diffusion-based generator, and constructed the MultiCities Dataset, a benchmark of 50,000 paired satellite-street-view images across five cities on five continents. Applying the framework to Kathmandu, Nepal improved usable street-view coverage by about 28%, showing a scalable way toward globally representative urban analytics.

Figure 1. Uneven street view imagery coverage in selected cities (top) and the GeoIdentity-Sat2Street generation framework (bottom).

The paper was published in ISPRS Journal of Photogrammetry and Remote Sensing

2 Visual auditing of built environments

How can SVI be used to audit built environments in a systematic and reproducible way? We conducted a systematic review of street view imagery-based built environment auditing tools, synthesizing the auditing dimensions (eye-level and sky-view angles), the detection and segmentation models behind them, and their applications across countries and research groups. Furthermore, we extended the auditing from 2D facades to 3D vertical cities by developing a 3D hedonic price model (3D HPM) that quantifies vertical urban features from SVI with machine learning for vertically developed cities.

Figure 2. An overview of the studies using different built environment auditing tools in different countries.

Figure 3. Illustrations of SVI sampling and the eye-level and sky-view angles (a)-(e).

The review was published in IJGIS, and the 3D HPM study was published in Habitat International

3 Eye-level greenery and multi-source urban analysis

Eye-level greenery often diverges from what aerial indicators suggest. We introduced a multi-perspective framework integrating the aerial perspective (NDVI) and the human-centric perspective (GVI) with urban morphological, socioeconomic, and topographic indicators, revealing how urban expansion reshapes the spatial relationships between urban green space and urban morphology in Guangzhou. We also fused SVI-based semantic segmentation with multi-source geospatial big data to assess the spatiotemporal dynamics of bikeability in Xiamen, evaluating daily safety, comfort, accessibility, and vitality of street cycling environments.

Figure 4. Research framework of the multi-perspective urban green space analysis.

Figure 5. The bikeability assessment: the study area of Xiamen Island, the proposed framework, the daily bikeability maps (December 21st–25th), and the field validation with street-level photos.

The Guangzhou study was published in Sustainable Cities and Society, and the bikeability study was published in International Journal of Applied Earth Observation and Geoinformation

4 Visual-semantic urban governance with large language models

Fine-grained fire hazards, such as cluttered wires or illicit ebike charging, are invisible to conventional POI-based risk indicators. We developed a visual-semantic risk indicator system that uses multimodal large language models (MLLMs) to extract fire-hazard features from street-view and remote-sensing imagery, and integrated these features into a geographically weighted XGBoost model for fire risk governance in Wuhan. The framework distinguishes urban areas that appear similar in static indicators but differ substantially in micro-scale hazard conditions, and scenario simulations show that interventions targeting informal practices in transitional areas produce the greatest reductions in fire risk.

Figure 6. Research framework of the MLLM-based urban fire risk governance.

The paper was published in Cities

Spatial Lifecourse Health

Thu, 01 Sep 2016 00:00:00 +0000

Spatial Lifecourse Health was proposed by my supervisor, Prof. Peng Jia in Lancet Planetary Health. It is an emerging framework that integrates life-course theory with spatial thinking to understand how places shape health across the life span, and it is now also applied to infectious diseases such as the COVID-19 pandemic. The top 10 research priorities in spatial lifecourse epidemiology and the ISLE-ReSt reporting standards have been published to guide this field. My PhD project applies Earth observation (Google Earth Engine), geospatial big data (street view image segmentation), and spatial statistical models to health geography and spatial epidemiology, covering obesity, infectious diseases, non-communicable diseases, and air pollution exposure.

1 Obesity and obesogenic environments

We conducted a series of systematic reviews in Obesity Reviews to determine all the obesogenic environmental factors of childhood obesity, covering the natural environment, green space, bike lanes, and speed limits, and further proposed a framework connecting built environments at different geographic contexts—residences, commute routes, and workplaces—with obesity.

The framework paper was published in The Lancet Global Health, and the residence–commute–workplace study was published in Cities

2 COVID-19 pandemic

The COVID-19 pandemic became a natural laboratory for spatial lifecourse health. I built Awesome of COVID-19, a website collecting COVID-19 research resources, and developed a COVID-19 shiny app for spatiotemporal visualization.

Figure 1. Snapshot of the website “Awesome of COVID-19”.

Figure 2. Snapshot of the COVID-19 shiny app.

We also raised awareness of reverse zoonoses (i.e., human–animal transmission of COVID-19), calling for up-to-date methods to improve the control, management, and prevention of cross-species transmission at both individual and regional/national levels.

Figure 3. COVID-19 natural infections of pet, zoo, and livestock animals as of 11 March 2021, mapped onto the number of confirmed human cases.

Trends论文献策疫情中被忽视的“人-动物”感染风险 | Cell Press论文速递

The paper was published in Trends in Ecology & Evolution, and the heterogeneous impacts study was published in Landscape Ecology

Our contributions to COVID-19 epidemic prevention and control were also recognized by the Health Commission of Tumot Left Banner, Inner Mongolia.

Figure 4. The certificate of outstanding contributions to COVID-19 epidemic prevention and control.

3 Health effects of air pollution exposure

Based on the daily 1-km PM2.5 mapping products, we evaluated the short-term health effects of PM2.5 exposure during the Spring Festival in 25 Yangtze River Delta cities using the attributable fraction (AF) and OD matrices, and further extended the exposure–health chain to long-term chemical constituents and diabesity risks in a national cohort, as well as real-time blood pressure responses measured by wearables.

Figure 5. The time variation and OD matrices of AF.

【建成环境与行为研究】春节期间的PM2.5污染短期暴露健康效应评估——以长三角地区25个城市为例 | 上海城市规划

The cohort study was published in The Lancet Regional Health – Western Pacific, and the wearable study was published in Journal of Urban Health

4 Built environment auditing and physical activity assessment

Street view imagery makes it possible to audit built environments at eye level. Through a systematic review of SVI-based built environment auditing tools, we synthesized the audited attributes, the detection models behind them, and their reliability across countries and research groups, and identified standardized tools (such as ANC and MAPS) as the most widely accepted auditing instruments.

Figure 6. An overview of the studies using different built environment auditing tools in different countries, (a) represents the number of these tools within different countries (in parentheses), (b) represents the number of these tools that have been applied by different groups and in various studies.

The review was published in IJGIS

We further developed a bikeability evaluation framework by fusing multi-source spatiotemporal big data, revealing the daily and hourly dynamics of cycling friendliness on Xiamen Island.

Figure 7. The proposed bikeability framework.

Figure 8. Average bikeability of Xiamen Island on December 21st, 22nd, 23rd, 24th, and 25th, 2020; the roads highlighted in red indicate lower levels of bikeability, whereas those in green indicate higher levels.

Figure 9. Average bikeability of Xiamen Island at 6:00, 7:00, 8:00, and 9:00 a.m.

《Int J Appl Earth Obs》发文：基于多源地理空间大数据的时空自行车可行性评估

The paper was published in International Journal of Applied Earth Observation and Geoinformation

5 Population and demographic mapping

Fine-scale population structure is the foundation of exposure and health assessment. We developed a contextualized geographically weighted neural network (CGWNN) for high-resolution population density mapping, and further estimated complete sex and age structures on 100 m grid cells by fusing census data, remote sensing imagery, and geotagged social media data with a multi-output random forest model.

Figure 10. Demographic composition data coverage and the distribution of geotagged social media data in Chengdu City.

The CGWNN paper was published in Applied Geography, and the demographic composition mapping was published in IJGIS