GOSAIC

Unprecedented environmental, social, economic and epidemiologic data is being generated on the SARS-CoV-2 pandemic. Our solution builds predictive geospatial models on disease impact to populations (initially SARS-CoV-2) using environmental, social and economic data as reference. Disease impact to populations (SARS-CoV-2 cases, COVID-19 deaths) is based upon the amount of factors (environmental, social, economic) that the population is experiencing as well as the rate and quantity of change in these factors. GOSIAC uses this data in machine learning to create a Global Open Source Artificially Intelligent Cloud computing (GOSAIC) spatial monitoring system. Over time this system will increase accuracy by learning about what data (environmental, social and economic) is most effective in predicting disease impact. GOSAIC can extrapolate what it has learned to areas with no disease data available or to emerging infectious disease we know little about. 

Environmental degradation, social changes, economic hardship and disease are interrelated pathways that impact health. Environmental, social and economic data can be used in monitoring systems to predict the impact of diseases on populations. Current monitoring systems do not use all available data. Without using this data, our monitoring systems have little predictive power (pixel level) and are largely generalized to administrative units (states). 

Our monitoring systems largely rely on the social determinants of health(e.g., socio-economic, demographic, and genetic conditions). However, this data is spatially coarse, infrequently updated, and costly to measure. Frequently updated, publicly available, fine scale data on environmental determinants of health (temperature, precipitation, air quality) are available but are not extensively used in monitoring systems.

Monitoring systems that have related the environment to health have done so primarily in a single static moment and are not dynamic. Without being dynamic, the monitoring systems don't learn from the data they are monitoring and do not keep up with rates and amounts of change. Without monitoring environmental, social, and economic data and how they are changing, transmission pathways and impact of diseases to populations are unable to be accurately predicted at a fine scale.

The GOSAIC codebase, RasterFrames, CircuitScape, and Maximum Entropy (MaxEnt), have been used in various settings worldwide. Astraea has used RasterFrames in code for creating their geospatial artificial intelligence (AI) platform, EarthAI. EarthAI has been used by big name companies for global AI analysis. MaxEnt has been used for predicting species presence based on environmental data and more recently for amounts of disease cases or deaths. Circuitscape uses data produced from MaxEnt to predict movement patterns (connectivity of habitats) of species, and more recently for disease transmission patterns.

By gathering data on predicted transportation pathways and expected disease impact to populations, we can start to monitor effectiveness of disease interventions. This is done by monitoring where interventions are occurring and comparing the actual amount of population to the predicted amount of population impacted by a disease. From this data the impact of interventions on health, economy and environment can be found.

According to the World Health Organization, 85 of the 102 major diseases are associated with the environment. A degraded environment disproportionately affects the most vulnerable populations, especially women and children. To ensure global health security and resilience, the environment must be incorporated into prediction models for disease impact. The information gained from GOSAIC will help government agencies, medical institutions, and supply chains (at the local to global level) to make decisions that decrease disease impact to populations. Effective environmental changes could fix at least 25% of the global disease burden. This was exemplified during the COVID-19 lockdown where a 63% reduction in NO2 over air pollution in Wuhan, China prevented an estimated 14,000 deaths.

The global community can increase their understanding of health through the non technical imagery portrayed on GOSAIC’s website. To remove language barriers and promote global accessibility, the written language used on the website is translated to the user’s language. Our global community connections (both current and future) allow us to understand the needs for GOSAIC. We expand the reach of our global community through social features (messaging, forum space), external events (hackathons) and feedback (user surveys and reviews).

GOSAIC’s monitoring system identifies vulnerable populations and provides an early warning for emerging infectious disease outbreaks. This system strives to constantly improve and incorporate best practices for the integration of genomic, behavioral, social and environmental data in surveillance of population health. The results displayed in the surveillance system are easy to read maps. By translating complicated technology and data to images, decision makers are better able to understand and take action on them. These maps depict the resistance of populations to diseases. From this resistance likely transmission patterns of different diseases are predicted. 

Evidence for the GOSAIC codebase is established, however continuous dynamic prediction of transmission pathways and disease impact to populations has not been done. A recent study demonstrated the effectiveness of using environmental data to predict disease impact to populations. This study was able to predict global infection rates of SARS-CoV-2 at a spatial resolution of 55 square kilometers and a maximum accuracy of 77.25% using four environmental data sources as reference. This disease impact data is then used for the prediction of transmission pathways. Disease transmission pathways (mainly for wildlife) have been predicted by using. Previous studies using these codes have produced a one time static image. This image can be made dynamic through the same data analysis methods, but running it continuously through our partner (Astraea) cloud computing codebase.

GOSAIC includes data that has not been used extensively before in monitoring systems (environmental) and analyzes this data in new ways (dynamic, real-time, AI) to improve the global knowledge of environmental and human health. Four key elements make GOSAIC different from other monitoring systems, breadth, resolution, real-time prediction, and action. The breadth of data used in GOSAIC includes more environmental determinants of health and new ways to analyze environmental data than any system before. Through using this data, the resolution of the imagery produced is finer scale than what other monitoring systems are able to achieve (such as animal based; bats, mosquitoes, etc. or human based; contact tracing, social determinants of health, etc.). This data is streamed and analyzed through cloud computing infrastructure allowing for real time prediction instead of static imagery. Dynamic imagery from GOSAIC shows specific factors (environmental, social, economic) that contribute to disease impact, transmission pathways and the amount of population impacted by diseases. This information, portrayed by GOSAIC, can lead to proactive responses for disease management.

GOSAIC is designed to scale with data and environmental/health challenges, as and when they happen. In year one GOSAIC’s focus is SARS-CoV-2. Initially one model (MaxEnt) trained on SARS-CoV-2 with environmental data as reference will be used. In years two to three, GOSAIC will scale monitoring to other emerging global health challenges. The predictive capacity of GOSAIC from year one will increase by using more data and models. In year one GOSAIC is expected to have a minimal accuracy level of 60% at a 30 square meter resolution. By year three the accuracy is expected to improve to at least 80% with a resolution of 15 square meters. 

GOSAIC will initially be launched in English, in years two to three GOSAIC will be promoted and translated to different languages. Effectiveness of ads will be tracked by social media and IP addresses accessing the platform. This info will further help target ads and increase the user base. Our revenue will be used to enhance GOSAIC and to implement interventions that have direct impact on populations. This impact can be increased by helping to locate interventions in areas that will directly increase the health of populations.  

For improvement of GOSAIC, we will measure the number of diseases covered, accuracy of disease impact predictions, pixel resolution of analyzed imagery, collaboration on coding, and linked data. In the first year GOSAIC will only cover SARS-CoV-2, expanding to other diseases in following years, at a resolution of 30 meters and at least 60% accuracy. GOSAIC’s code ,used to analyze data for SARS-CoV-2 and other diseases, will improve with new code commits. Our project partner’s code, RasterFrames, currently has 1,402 code commits. Initially disease data will be at the country level, however the spatial scale of this data will improve as new publicly available data becomes linked.

For monitoring the user base of GOSAIC, we will measure number of users, location of users, and social media followers. Direct tangible impact of GOSAIC to the global community is measured through money spent on interventions, CO2 sequestered, and eventually economic impact of interventions (through incorporating economic data into GOSAIC). Astraea currently has 1,770 followers on Twitter globally. No money has been spent on interventions and there has been no tracking of CO2 sequestered yet. 

Dr. Aris Persidis, Big Data Biotech, Biovista

Dr. Suzanne D. Vernon, Virologist and Clinical Research Scientist, Independent Advisor

Dr. Sepul Barua, Environmental Economist 

Dr. Anita Wreford, Climate Change Expert

Dr. Garry McDonald, Market Economics Ltd.

Dave Yoken, Spatial Data Intelligence, Astraea 

T.R. Price, Environmental Spatial Analyst, Treetable LLC

Our team has experts in the environment, infectious disease, biochemistry, computer sciences, and geographic information systems. Astraea is represented as a member partner on our team through David. Astraea’s EarthAI platform is able to be utilized in this project guided by professional modeling experience from Garry and T.R.. Garry helped in the creation of one of the world’s leading ecosystems for spatial analysis of cascading disasters, MERIT. This is complemented by T.R.’s experience in spatial evaluation of ecosystem services through InVEST. 

This knowledge of how the environment is tied to human health is furthered by that of Sepul, Anita, and Suzanne. Sepul works as an international consultant for ecosystem services and has ties to the Food and Agricultural Organization. Anita is an expert in climate change and co-authored the International Panel for Climate Change report on land use and climate change. Suzanne specializes in infectious causes of chronic disease with current focus on post-acute sequelae of SARS-CoV-2 infection.

In order to make use of this experience a structure, artificial intelligence, must be developed. Aris is a serial entrepreneur in the fields of big data and biotech. In several of his businesses he helped to develop revolutionary AI, such as Vizit. 

Our leadership team is composed of experts from 3 different countries. We will further encourage this international composition throughout the development of GOSAIC. We offer a flexible work environment and are committed to cultural inclusion. Currently we are drafting a diversity mandate and always work for equitable treatment of everyone. 

The key barrier to our solution is the need for upfront investment. Setting up GOSAIC requires significant data collection, developing tools for data and image processing and classification as well as developing the online platform, and coding. These all are time and resource intensive activities. Our team consists of a group of researchers and professionals that do not have sufficient resources at their disposal for starting up GOSAIC. We would need to hire additional personnel for this purpose. The grant will help us finance the upfront investment needed during the start-up phase of GOSAIC. 

GOSAIC needs technology experts for maintaining it and making sure it is secure. The founding members of the Trinity Challenge have significant technical expertise that would allow for superb security and maintaining GOSAIC. 

Moreover, we expect to gain new ideas, expand our network and connect with potential GOSAIC stakeholders through the Trinity Challenge platform. We believe these would be beneficial for continuously improving our solution and sustaining it beyond the start-up phase.

Financial resources are needed in order to hire personnel.

Palantir, John Hopkins, Healthmap and ProMed have experience in global disease data aggregation for fine scale administrative boundaries, this data would supplement large administrative unit data. This fine scale data could be supplemented by contact tracing, i.e. Google. Search data from Google and social media data from Facebook could be used in conjunction for early detection of sick populations and possible outbreaks. For transmission of diseases, mobility data from Google, Cuebiq and MapBox would be an important input. Tsinghua has demonstrated expertise in spatial analysis with their Land Use Land Classification classifier. This expertise can be furthered through other open source GIS organizations, such as QGIS. Hyperledger and the Linux Foundation would help to ensure open source standards are met and that we are implementing the latest technology. The infrastructure is currently hosted on Amazon Web Services, but could be migrated to the Microsoft Azure interface. The partnerships formed with these organizations can help in the maintenance and security of GOSAIC.

Our health is dependent on our environment and from the start GOSAIC will support nature-based interventions. Nature-based interventions improve over time, unlike most human created interventions. For example, as a tree grows its ability to filter the air improves whereas a face mask filtering ability degrades. 

Implementing interventions can improve not only our health but also our productivity. The entirety of GOSAIC’s operations is planned to be carbon (CO2) negative. This is done by partnering with organizations to pay people in developing countries to plant and maintain trees. By decreasing CO2 levels (in areas with high carbon dioxide levels) higher levels of productivity (cognitive functions) are able to be experienced. 

GOSAIC’s monitoring system, eases the burden for monitoring disease impact on populations. Easing this burden allows for more resources to go towards proactive interventions, such as nature-based interventions (or other data backed interventions), that improve the health and productivity of everyone. 

GOSAIC is especially important for populations (especially women and children) in developing countries, and other vulnerable areas. These areas are disproportionately impacted by diseases, often do not have the resources for monitoring diseases, and are limited in the resources they can put towards interventions. 

Our health is dependent on our environment and from the start GOSAIC will support nature-based interventions. Nature-based interventions improve over time, unlike most human created interventions. For example, as a tree grows its ability to filter the air improves whereas a face mask filtering ability degrades. 

Implementing interventions can improve not only our health but also our productivity. The entirety of GOSAIC’s operations is planned to be carbon (CO2) negative. This is done by partnering with organizations to pay people in developing countries to plant and maintain trees. By decreasing CO2 levels (in areas with high carbon dioxide levels) higher levels of productivity (cognitive functions) are able to be experienced. 

GOSAIC’s monitoring system, eases the burden for monitoring disease impact on populations. Easing this burden allows for more resources to go towards proactive interventions, such as nature-based interventions (or other data backed interventions), that improve the health and productivity of everyone. 

GOSAIC is especially important for populations (especially women and children) in developing countries, and other vulnerable areas. These areas are disproportionately impacted by diseases, often do not have the resources for monitoring diseases, and are limited in the resources they can put towards interventions. 

Our health is dependent on our environment and from the start GOSAIC will support nature-based interventions. Nature-based interventions improve over time, unlike most human created interventions. For example, as a tree grows its ability to filter the air improves whereas a face mask filtering ability degrades. 

Implementing interventions can improve not only our health but also our productivity. The entirety of GOSAIC’s operations is planned to be carbon (CO2) negative. This is done by partnering with organizations to pay people in developing countries to plant and maintain trees. By decreasing CO2 levels (in areas with high carbon dioxide levels) higher levels of productivity (cognitive functions) are able to be experienced. 

GOSAIC’s monitoring system, eases the burden for monitoring disease impact on populations. Easing this burden allows for more resources to go towards proactive interventions, such as nature-based interventions (or other data backed interventions), that improve the health and productivity of everyone. 

GOSAIC is especially important for populations (especially women and children) in developing countries, and other vulnerable areas. These areas are disproportionately impacted by diseases, often do not have the resources for monitoring diseases, and are limited in the resources they can put towards interventions. 

GOSAIC includes data that has not been used extensively before in monitoring systems (environmental) and analyzes this data in new ways (dynamic, real-time, AI) to improve the global knowledge of environmental and human health. Four key elements make GOSAIC different from other monitoring systems, breadth, resolution, real-time prediction, and action. The breadth of data used in GOSAIC includes more environmental determinants of health and new ways to analyze environmental data than any system before. Through using this data, the resolution of the imagery produced is finer scale than what other monitoring systems are able to achieve (such as animal based; bats, mosquitoes, etc. or human based; contact tracing, social determinants of health, etc.). This data is streamed and analyzed through cloud computing infrastructure allowing for real time prediction instead of static imagery. Dynamic imagery from GOSAIC shows specific factors (environmental, social, economic) that contribute to disease impact, transmission pathways and the amount of population impacted by diseases. This information, portrayed by GOSAIC, can lead to proactive responses for disease management.

GOSAIC’s monitoring system identifies vulnerable populations and provides an early warning for emerging infectious disease outbreaks. It quantifies the number of people expected to be impacted by an illness. This number can be used to determine the amount of medicine needed for that area. The analyzed data forms an information repository where users can select dates for data and view the analysis. GOSAIC is automated to the fullest extent possible. This system strives to constantly improve and incorporate best practices for the integration of genomic, behavioural, social and environmental data in surveillance of population health. The results displayed in the surveillance system are easy to read maps. By translating complicated technology and data to images, decision makers are better able to understand and take action on them. These maps depict the resistance of populations to diseases. From this resistance likely transmission patterns of different diseases are predicted. As GOSAIC learns more about how monitored data impacts health, the highest risk spillover zones are able to be determined. These zones can then be closely monitored to determine when a spillover is likely to happen. Interventions can be suggested to improve the resistance of populations to spillovers and diseases. Effectiveness of interventions can be measured by the information monitored by GOSAIC.