COVID-19 resource page for AI researchers
Machine learning techniques are already playing a key role in assisting research into COVID-19. Here we provide a living document where we will add links to resources as and when we become aware of them. We hope this will be useful for AI researchers wishing to contribute their expertise to projects and collaborations.
Conferences, forums, webinars and collaborations
CLAIRE task force on COVID-19
CLAIRE (Confederation of Laboratories for Artificial Intelligence Research in Europe) has offered governments and health systems the free support of its network of AI experts. Here is a list of existing initiatives (projects, funding calls, hackathons, open datasets or tools, etc).
Sign up as a volunteer in CLAIRE efforts against COVID-19.
Submit information to CLAIRE about new COVID-19 initiatives, news items and datasets, etc.
CLAIRE Task Force on AI and COVID-19: Results and next steps
A webinar organised by CLAIRE took place on 15 July 2020. You can watch it here.
ELLIS against COVID-19 online workshops
Hosted by the ELLIS society, these workshops cover outbreak prediction, epidemiological modelling, drug development, viral and host genome sequencing, and health care management. The events are held on Wednesdays, 13:30 – 16:00 (CEST). You can watch the past workshops from these links: 1 April, 15 April, 22 April, 6 May.
COVID-19 and AI: A Virtual Conference
This event was hosted by the Stanford Institute for Human-Centered Artificial Intelligence, Stanford University and was held on 1 April 2020. Topics addressed included: AI applications in diagnostics and treatment, epidemiological tracking and forecasting of the spread of the virus, information and disinformation, and the broader human impact of COVID-19 and pandemics in general on economies, culture, government, and human behaviour. You can watch the event here.
CECAM (Centre Européen de Calcul Atomique et Moléculaire) are running a series of webinars on “CoVid-19: challenges and responses in simulation, modeling and beyond”
The first webinar was held on 21 April and you can view the talks from this, and the subsequent webinars, here.
Join the AI-ROBOTICS vs COVID-19 initiative of the European AI Alliance
The European Commission has launched an initiative to collect ideas about deployable Artificial Intelligence (AI) and Robotics solutions as well as information on other initiatives that could help face the ongoing COVID-19 crisis. The initiative aims to create a unique repository that is easily accessible to all citizens, stakeholders and policymakers and become part of the joint European response to the outbreak of COVID-19. Click here to find out how to submit your ideas.
GOVLAB have created a living repository – add information about your projects
The repository is part of a call for action to build a responsible infrastructure for data-driven pandemic response. It serves as a source for data collaboratives seeking to address the spread of COVID-19 and its secondary effects.
The European Commission hosted a pan-European hackathon
This took place on 24, 25, 26 April with the aim of connecting civil society, innovators, partners and investors across Europe in order to develop innovative solutions for coronavirus-related challenges. You can see the results here. Here is a short video introducing the project.
Join Montreal AI on Slack
They team are inviting people to join them to collaborate on matters of importance (including the COVID-19 outbreak).
COVID-19 forum hosted by fast.ai
This is a forum for discussions that are data-driven, technical and/or practical. Share and join projects for building ventilators, finding reagents for testing, improving contact tracing, public policy discussions and communications, etc.
UK call for modellers to support epidemic modelling
The Royal Society is coordinating a scheme to allow those with modelling skills (including data science) to contribute to current UK efforts in modelling the COVID-19 pandemic. More details, including the online form can be found here.
Call for COVID-19 rapid response data science taskforce
DECOVID is a collaboration between The Alan Turing Institute, HDR UK and founding Trusts and academic partners at University Hospitals Birmingham/University of Birmingham and University College London Hospitals/University College London. DECOVID aims to address urgent questions defined by clinical staff in order to produce rapid insights which are actionable on the frontline at a local level. This is a joint call for teams of researchers as well as individual researchers to join DECOVID in addressing urgent questions. Teams will rapidly trial tested, robust and reproducible data science and machine learning methods on the new national data resource of COVID ICU data, to create a near real-time, human-in-the-loop analytics platform and information console to support acute care management and clinical decision making for COVID-19. The form with detailed information can be found here.
Join Beat19.org – a real-time initiative to gather and analyze data, and get the results to those who need it
Sign up to complete brief daily email surveys and be immediately alerted to actionable findings regarding COVID-19.
Datasets and code
COVID-19 Open Research Dataset (CORD-19)
An open-source, machine-readable collection of scholarly literature covering COVID-19, SARS-CoV-2, and the Coronavirus group. This free resource contains over 29,000 relevant scholarly articles, including over 13,000 with full text.
Johns Hopkins University dashboard of COVID-19 cases
The data for this resource is on GitHub so others can copy and modify as required.
Database of chest x-rays from University of Montreal
The team are building a database of COVID-19 cases with chest X-ray or CT images. They are looking for COVID-19 cases as well as MERS, SARS, and ARDS.
COVID-19 Korea dataset
This is a comprehensive dataset of all the available data on the disease from Korea.
Provided by the National Center for Biotechnology Information, USA.
The COVID-19 Vulnerability Index (CV19 Index)
This repository contains the source code, models, and example usage of the COVID-19 Vulnerability Index (CV19 Index). The CV19 Index is a predictive model that identifies people who are likely to have a heightened vulnerability to severe complications from COVID-19.
This dataset contains the Tweets of users who have applied the following hashtags: #coronavirus, #coronavirusoutbreak, #coronavirusPandemic, #covid19, #covid_19
A further tweet dataset may be found here.
Code for modelling estimated deaths and cases for COVID19
This is available on the Imperial College London GitHub page.
Neural Covidex is an AI-powered search engine designed for the CORD-19 dataset
It was created by teams from University of Waterloo and New York University.
Funding for COVID-19 research
The EU have issued a second special call for expressions of interest for funding to aid the COVID-19 response
The deadline to send in your expression of interest is 11 June 2020. Find out more information here. You can also watch a recording of the live-streamed video giving more information about the funding process here.
The C3.ai Digital Transformation Institute has issued a call for proposals to receive funding for research into AI techniques to mitigate COVID-19 and future pandemics
C3.ai DTI is a new research consortium established by C3.ai, Microsoft Corporation, the University of Illinois at Urbana-Champaign (UIUC), the University of California, Berkeley, Princeton University, the University of Chicago, the Massachusetts Institute of Technology, Carnegie Mellon University, and the National Center for Supercomputing Applications at UIUC.
Up to $5.8 million in awards will be funded from this first call, ranging from $100,000 to $500,000 each. In addition to cash awards, C3.ai DTI recipients will be provided with significant cloud computing, supercomputing, data access, and AI software resources and technical support provided by Microsoft and C3.ai. The first call for proposals is open now, with a deadline of May 1, 2020. You can find out have to submit your proposal here.
Get funding for ideas that address COVID-19 (UK only)
UK Research and Innovation is inviting proposals for short-term projects addressing and mitigating the health, social, economic, cultural and environmental impacts of the COVID-19 outbreak.
Information and articles
COVID-19 information guide from Gerstein Science Information Centre, University of Toronto Library
This comprehensive set of webpages provides information and news about the virus and resources for researchers
Mapping the landscape of artificial intelligence applications against COVID-19
This arXiv article presents an overview of recent studies using artificial intelligence to tackle many aspects of the COVID-19 crisis at different scales including molecular, medical and epidemiological applications.
Interactive website that gives a ten-day forecast, by country, on likely numbers of COVID-19 cases
This was developed by a team from the University of Melbourne. The code is available here.
Computing power and resources
Donate some of your computing power to folding@home to help run simulations
Download folding@home and the software runs while you are doing other things.
Get free computing power for COVID-19 research
Rescale, Google Cloud and Microsoft Azure announced a new program that immediately offers high performance computing resources at no cost to teams working on COVID-19.
Sign up here.
Element AI provides a search platform to assist in finding information from research papers
This natural language web platform is designed to help with searches of structured and unstructured information and the initial version has been configured to work with the COVID-19 Open Research Dataset (CORD-19) corpus.
The Australian National Computational Infrastructure (NCI) and Pawsey Supercomputing Centre (Pawsey) have joined forces to offer additional computation and data resources
This support is available to the national and international research community. Visit their page to find out how to apply.
Research tasks, classes and challenges
MIT and J-Clinic announce an open task: fighting secondary effects of COVID-19
The task is to predict a target compound’s property from its molecular structure. You can find out more here. The open datasets can be accessed from the website or from GitHub.
Stanford University are organising a “data science and AI for COVID-19” project class
This started on 10 April and they plan to make a lot of the information public.
If you have any suggestions for resources or links we should add to the page then please get in touch at aihuborg[at]gmail.com.