$500K prize anticipated for 2 IIT Grads, builds A.I. Elixir to Predict Future Pandemic

A year has passed since all of us were locked indoors due to the ongoing SARS CoV-2 pandemic that wreaked havoc on our lives. Although most of us have started accepting our fate as ‘new normal the pandemic is showing no signs of slowing.

Indeed this pandemic has been a learning experience for most of us- doctors, engineers, scientists, businesses, and commoners. Nobody wants to relive this agony. The government and other institutions in the world are getting prepared, in case the next pandemic messes things up (instead of remaining ignorant like the COVID-19 pandemic).

Pandemic hit the Nation pretty hard, streets were almost deserted.

PIC CREDIT: BUSINESS STANDARD

Likewise, a duo of Indian-origin AI experts is trying to bridge the gap between ignorance and vigilance.

Two IIT Kanpur Graduates, Mudit Jain, an AI Engineer at Google, and Biplab Deka, Ph.D. from the University Of Illinois, UC have teamed up to create an Artificial Intelligence-based solution that can predict COVID-19 cases accurately and prescribe Intervention plan solutions.

Keeping in mind different locations having different patterns, they’ve come up with a scalable solution to this problem. They will create separate A.I. modules for each region to predict COVID-19 time-series in those regions.

An SEIR Based Solution:

The two ignited minds built this solution on a traditional epidemiological technique called SEIR Model and extended it with A.I. to make it dynamic, instead of the traditional static approach. Mudit stated the acronym SEIR meant, Susceptible-Exposed-Infected-Recovered. It’s a 90-year old modeling technique that models the no. of people in each category using differential equations and static assumptions.

 

This is one of the pre-designed A.I. models predicting the trends of the endemic in China.

SEIR Model: A.I. Prediction of the Endemic Trends

PIC CREDIT: Google Images

 

Mudit further added, “We extended the SEIR model using A.I. and made the model dynamic, i.e., location and time-dependent. So those static assumptions have been replaced by dynamic data-dependent outputs of A.I. models, thus making the whole model accurate.”

For training the models, they made use of existing publicly available features like demographics, health infrastructure, and economic indicators.

Data Sourcing From Google Maps:

They also accessed the Google Maps mobility index which is collected from Android phones worldwide in an anonymized and privacy-secured way. This measures population movement changes in public areas such as parks, workplaces, groceries, railway stations, airports, etc.

Another unique feature they used was Google’s COVID-19 symptom search trends data that is similarly aggregated anonymously and contains the popularity of Google search queries related to COVID-19 symptoms.

Is this solution Credible? Will it Resolve Anything?

Mudit shared how their A.I.-based implementation can offer a more accurate prediction of cases in the future. “By using location-wise (country/state-wise) model predictions, healthcare policymakers can more accurately estimate the no. of cases in the coming months. That can help them proactively decide preventative measures to take such as travel restrictions, varying degrees of lockdowns, health awareness advertising campaigns, etc.”

“They can also prioritize vaccination programs to target more vulnerable regions first. The model being dynamic, will also adapt to the impact of these measures and update predictions as new data arrives.”

COVID-19 Rapid Test PIC CREDIT: Google Images

Ability to Predict Future Pandemics:

Mudit further explained that pandemics may be caused by miscellaneous sources, but they are not different from one another in terms of outcome.

He added, ” Mathematically, all past, current, and future pandemics follow a similar process and equations. This is why all public health experts were talking about ‘flattening the curve‘ and ‘R-nought values‘ even before COVID-19 started to spread widely”.

“Since the underlying mathematical modeling for each pandemic is the same, even though the virus’s biological mechanism may be quite different, this A.I.-augmented SEIR technique can be used for future modeling too”.

 

The $500,000 Prize:

The A.I. Elixir built by Mudit and Biplab is one of the finalists for the Cognizant XPrize Pandemic Response Challenge. Likewise, Victory in this would award them with a $500,000 cash prize. They have managed to reach the finals based on the accuracy of their predictions, the application of novel A.I. techniques, and the ease of interpretability of their predictions.

ALSO READ:

After Simpson’s ‘Covid Outbreak’, 1997 Archie Comic predicted the ‘New Normal’ of 2021; Blew the Internet.

‘India is very Unhappy’; Ranks 139 out of 149 countries, Finland Tops the list: Happiness Report 2021