How Geospatial Data and Technologies can Help in Disease Prevention and Control

GeoAI or Geospatial Artificial Intelligence is an emerging scientific discipline that combines innovations in spatial science, Artificial Intelli­gence methods in Machine Learning, Big Data mining, and high-performance computing to extract knowledge from spatial Big Data.

GeoAI is increasingly being used to model and capture the environment around us, linking locations in which we live and work, or people/elements we interact with, to explore their potential role in influencing health outcomes. There is also extensive research into GeoAI being used for hypothesis generation, conducting new data linkages and predicting disease occurrence.

With the development of mobile technology, the users’ locations can be identified by mobile phone, which means signalling data can obtain personnel location information to track personnel trajectories. The disease prevention and control organizations can analyze close contact groups based on the travel information of diagnoses. Thus they can quickly find suspected patients and close contacts through data retrospective analysis, which is the so-called “contact tracking” and helps to quarantine and cut off the source of infection in time, explains Zhang.

Location analytics provide useful tools to model behaviors and inform actions. From maps that analyze the genetic profile of the virus as it spreads from place to place to AI techniques that make sense of human movement data, we can enhance our understanding of viral transmission, determine if public health recommendations are being followed and predict whether travel bans and other measures will quell the spread of disease, adds Dr. Geraghty.

There are examples where GeoAI was used in infectious disease modelling or prediction of disease occurrence and for disease surveillance. For instance, Deep Learning recurrent neural networks were used for real-time influenza fore­casting at regional and city spatial scales in the US using spatial Big Data on Google Flu Trends and climate data (such as precipitation, temperature, sun exposure) from the National Climatic Data Centre. Geotagged tweets were analyzed against the CDC influenza-like illness (ILI) dataset to predict real-time regional ILI in the US using an artificial neural network (ANN) optimized by an artificial tree algorithm.

China earlier used Machine Learning to accurately forecast dengue outbreak in 2014 using climate data, weekly dengue fever cases, and research queries on the Chinese Internet search engine Baidu.

Advancements in Artificial Intelligence have also seen a growing interest in real-time syndromic surveillance based on social media data in recent years. Deep Learning algo­rithms can be applied to Twitter data to detect illness outbreaks and then to build up and display information about these outbreaks, including relevant news articles, to provide situational awareness. In the US, this has demonstrated an ability to detect symptoms for Influenza-like illness, which were then confirmed from the CDC Morbidity and Mortality Weekly Reports (MMWR). There is further research onto improve on this sur­veillance system to incorporate disease-spe­cific information (e.g., mode of transmission) to enhance disease forecasting accuracy.

When a major epidemic comes, the impact of panic on social operation may exceed the viral disease itself. To this end, it is necessary to track and evaluate the spatial spread of social emotions by analyzing massive social media data. For instance, as Li of SuperMap says, when facing an epidemic, public behavior might be irrational, highly infectious, and conformable. It is required to build a knowledge base of epidemic-related emotions and to dig out the dynamic evo­lution of public opinion in time, space and semantics aspects from social media.

By using Internet social data as a data source, the public topic categories from social data related to the epidemic can be obtained based on the construction of topic extraction and sentiment classification framework by topic models and Machine Learning methods. Based on the complex networks, the changing network of public topic can be built. Also, by using the network model, the public dynamic changes in topical emotions can be character­ized. These outcomes contribute to the reveal of the temporal, spatial and semantic distribution characteristics and evolution patterns of public topic views under the COVID-19, adds Zhang.

According to Luis Sanz, CEO, CARTO, innovative statistical methods and computa­tional tools can be used for public health sur­veillance including spatio-temporal models for disease risk prediction, cluster detection, and travel-related spread of disease, which can further inform strategic policy in reduc­ing the burden of diseases.

“We are seeing an increasing trend of using geospatial tools for prevention and con­tainment. An example of using geospatial for analysis and not just visualization is this risk analysis carried out by researchers in Spain, Brazil and the US,” he points out.

The COVID-19 Map of Propagation Risk in the three countries aims to show the results of the estimated epidemic risk right down to the municipal level by modelling the epidemic spread which takes into account the recurrent mobility patterns (commuting) among municipalities. Incidentally, update of the map risk for Spain has been suspended temporarily due to unavailability of real mobility data following the declaration of a state of alarm in the country.

The AsistenciaCovid19 app is an interesting example. While the primary and initial aim of the app is to reduce the pressure on emergency systems and track the status of symptoms when people are taking care of themselves at home, it also provides a method to understand the pandemic from a spatio-temporal perspective. Since there is a location element to the data being collected, the local authorities can visu­alize infections on an interactive map and per­form geospatial analysis to determine high risk areas. Governments can see how symptoms change over time and by location, allowing them to act faster in certain hotspots.

Kumar points to the emergence of wear­ables and connected devices in the past few years that are capable of collecting a reasona­ble amount of individual health information such as heart rate patterns, sleeping patterns, etc. “Integrating this data into GIS technolo­gies could help healthcare workers to uncover long term geographic trends in health of cer­tain demographics or individuals living within certain regions, thus opening new realms of healthcare research and providing insights not previously attainable,” he believes.