C3 AI COVID-19 Grand Challenge

Change the World.
Be Bold.
Be Courageous.

The C3 AI COVID-19 Grand Challenge

In the C3 AI COVID-19 Grand Challenge, developers, data scientists, students, and creative minds around the world developed meaningful data-driven insights to inform decision makers and change how the world is fighting this pandemic.

Participants used the C3 AI COVID-19 Data Lake as one of the key sources of data to produce new and innovative data science projects that help combat the COVID-19 pandemic.

Globe

43

Countries

Participants

777

Participants

Congratulations to the Winning Teams

First Place

$100,000

Combined Diagnostics, Modeling Population Heterogeneity
by Providing Personalized COVID Diagnostics

Team: Freddy Bunbury, Nina Miolane, Claire Donnat

Second Place

$25,000

Confidence in the ‘Cures’? Assessing the Conservation Level of Therapeutic and Vaccine Targets of COVID-19 by Providing Personalized COVID Diagnostics

Team: Yuanfang Guan and Manqi Zhou

A Framework for Forecasting Dynamics of COVID-19 and Co-Evolving Diseases

Team: Alexander Rodriguez, Nikhil Muralidhar, Bijaya Adhikari, Naren Ramakrishnan, B. Aditya Prakash

Third Place

$12,500

Analysis of COVID19 Policies’ Impact on the Population and the Economy

Team: Niti Wattanasirichaigoon, Napat Ratanakul, Apiwat Jaroonpol

The Center on Stochastic Modeling, Optimization, and Statistics (COSMOS)

Team: Victoria Chen, Yuan Zhou, Nilabh Ohol, Alireza Fallahi, and Amith Viswanatha

COVIDProf: An Intelligent Q&A Chatbot Tuned to Deliver Scientific Research and Data Related to COVID-19 in Natural Language

Team: Varsha Rao, Joy Yeh

Structural Modeling of COVID Spread in Relation to Human Mobility

Team: Rezwana Rafiq, Tanjeeb Ahmed, Md Yusuf Sarwar Uddin

C3 AI COVID-19 Data Lake

Access unified, analysis-ready COVID-19 data – free of charge.

Learn More

Third Place

Team members: Rezwana Rafiq, Tanjeeb Ahmed, Md Yusuf Sarwar Uddin

Structural Modeling of COVID Spread in Relation to Human Mobility

The novel coronavirus pandemic is ravaging the US, causing one infection every three seconds and claiming one life every two minutes. Humans are the sole carrier of this virus; human mobility is a leading factor in COVID-19 spread. This study models the underlying causal relationships between various human mobility indicators – including trips, distance traveled, staying at home, and social distancing – and COVID-19 spread to help policymakers take necessary mobility-oriented actions to control the spread.

Third Place

Team members: Varsha Rao, Joy Yeh

COVIDProf: An Intelligent Q&A Chatbot Tuned to Deliver Scientific Research and Data Related to COVID-19 in Natural Language

Humanity is suffering not only from the COVID-19 pandemic but also from misinformation that amplifies the pandemic’s harm. We present COVIDProf, an intelligent Q&A chatbot tuned to deliver scientific research and data related to COVID-19 in natural language. It understands casual language inquiries and responds with scientific information extracted from the CORD-19 research paper dataset, along with other sources of data including infection and death statistics. Our goal is to make finding and understanding relevant COVID-19 knowledge easier to prevent to prevent the harm that ignorance can cause.

Third Place

Team members: Victoria Chen, Yuan Zhou, Nilabh Ohol, Alireza Fallahi, and Amith Viswanatha

The Center on Stochastic Modeling, Optimization, and Statistics (COSMOS) COVID-19 Decision Analytics

The end is not in sight for this pandemic and we should stop relying on hope. Nations are facing two conflicting objectives: To keep severe illness counts low and to revive economies and livelihoods for billions of people. We focus on “key contacts” to protect individuals at higher risk from the rest of the population and optimize usage of interventions, including testing, personal protective equipment, and vaccines. Findings indicate that we should not close schools and we cannot isolate all high-risk individuals. Instead, we must identify key contacts and enable higher protections for them, especially in lower-income neighborhoods.

Third Place

Team members: Niti Wattanasirichaigoon, Napat Ratanakul, Apiwat Jaroonpol

Analysis of COVID19 Policies’ Impact on the Population and the Economy

Government policies during the COVID-19 pandemic contributed to changes in lifestyles and societies. We introduce a policy evaluation method which assumes that, after a policy is issued, it changes the trend of various parameters that relate to the pandemic; comparing of pre- and post-trend measures can calculate the impact score of a policy. The analysis shows that each policy causes a different impact on the pandemic. Evaluating impact scores can aid government decision-making processes and lay a foundation for developing new trend analysis methods.

Second Place

Team members: Alexander Rodriguez, Nikhil Muralidhar, Bijaya Adhikari, Naren Ramakrishnan, B. Aditya Prakash

DeepOutbreak: A Framework for Forecasting Dynamics of COVID-19 and Co-Evolving Diseases

We propose DeepOutbreak, a holistic approach for modeling the progression of COVID-19 and symptomatically similar co-evolving influenza-like illnesses (ILI). DeepOutbreak contains a forecasting model for COVID-19 hospitalizations and mortality. During the pandemic, people afflicted with symptomatically similar illnesses like influenza change how and when they seek healthcare. This has caused unseasonal increases in hospital traffic for people suffering from ILI, which traditional models fail to forecast. Therefore, we propose having another model which leverages traditional models and COVID-19 signals for effectively ILI forecasting in this novel scenario. DeepOutbreak enables policymakers to design effective healthcare response strategies.

Second Place

Team members: Yuanfang Guan and Manqi Zhou

Confidence in the ‘Cures’? Assessing the Conservation Level of Therapeutic and Vaccine Targets of COVID-19 by Providing Personalized COVID Diagnostics

The virus sequence is evolving continuously, negating some of the epitopes and mRNA sequences as candidates for antibody and vaccine development. Through a combinatory approach exploring SARS-CoV2 genomic sequence variations, protein sequences, and literature reported epitopes, we report the conservation level of each mRNA sequence and the epitopes. Epitopes on average maintained 93% fidelity among the population (meaning an average 7% missing rate), while the highest examples can reach 99.89% fidelity in the population, pointing to targets we should prioritize for vaccine and drug development and deployment.

First Place

Team: Freddy Bunbury, Nina Miolane, Claire Donnat

Combined Diagnostics, Modeling Population Heterogeneity by Providing Personalized COVID Diagnostics

When will we be able to safely travel again or gather together in public? In short, when we can determine that the risks are acceptably low. Unfortunately, most models that predict the risk of infection at public gatherings fail to truly capture the heterogeneity in the infectiousness and susceptibilities of the people present, yet the spread of infectious diseases are dominated by tail events. Our Bayesian Stochastic Expectation-Maximization method provides personalized estimates for each individual’s infectious and susceptibility statuses by combining data on location, behavioral factors, symptoms, and various COVID-19 test results to inform individual decisions and public policy.

C3 AI Applications

C3 AI Platform

C3 Generative AI

Get Started with a C3 AI Pilot

Generative AI

Publications

What is Enterprise AI

Customer Viewpoints

Machine Learning

Blog

C3 AI Live

Glossary

Developer Portal

Company

Leadership

Partners

C3 AI DTI

Investors

Events

Careers

C3 AI Fellows

Change the World.

Be Bold.

Be Courageous.

The C3 AI COVID-19 Grand Challenge

43

Countries

777

Participants

Congratulations to the Winning Teams

First Place

$100,000

Second Place

$25,000

Third Place

$12,500

C3 AI COVID-19 Data Lake