AI and big data joins effort to predict deadly disease outbreaks

Editor’s Note: This feature is part of Tomorrow’s Hero, a series profiling young innovators transforming the world for a brighter future. Discover their stories here.

CNN  — 

Rainier Mallol’s journey started with an infection. Dengue fever, ripping through mother’s body when he was 14 years old.

Fever, headaches, vomiting, muscle and joint pains, transmitted by a mosquito bite in his native Dominican Republic. All too common in the tropics – and with cases rising dramatically in recent decades according to the World Health Organization – Dengue can become life-threatening. Although there is a vaccine, there is no specific treatment once contracted.

Rainier Mallol holds up a sample slide.

“I felt powerless,” Mallol reflects, 12 years later. His mother was one of an estimated 390 million Dengue infections every year; one of an estimated four billion people at risk of Dengue infection globally.

But what if this big data could be harnessed? If we can track the spread of previous disease outbreaks, can we find patterns to predict where they’ll happen next?

These were the questions on Mallol’s mind in 2015 when he took part in a program at the NASA Research Park in California. Eighty people from around the world were invited by the Singularity University (a Silicon Valley thinktank and business incubator offering education programs) to workshop an idea: “a solution that would save billions (of lives) in the next 10 years,” says Mallol.

They came from different backgrounds, among them journalists, health professionals, writers and film producers. Mallol was a computer engineering graduate and hit it off with Malaysian public health specialist Dr Dhesi Raja on their first day on the program.

The two talked about dengue and the inability to forecast outbreaks, and conceived an algorithm-based approach to monitor the disease. Mallol created a basic artificial intelligence software in a week, feeding it Dengue statistics and asking it to predict three months ahead.

“We waited about 10 weeks then compared the results,” he says. “We obtained about 81% accuracy.”

It was the first steps towards co-founding their company, Artificial Intelligence in Medical Epidemics (AIME), which today claims an improved average accuracy of 86.4%. Their company’s Dengue Outbreak Prediction platform now supplies the Malaysian government and regional governments in Brazil and the Philippines with insights to manage and curb outbreaks.

rainier mallol headshot
Predicting the next deadly outbreak
02:30 - Source: CNN

“What we do is we analyze data from the past year, past months, past days of dengue cases,” Mallol explains. Plotting these on maps, they add other factors. “We have a system that obtains over 276 variables,” he adds, “weather variables, geographical variables (and) socio-economic variables.”

“(Governments) can actually take pre-emptive actions,” says Mallol, using fumigation of zones likely to be affected as an example. Authorities can easily break down an ongoing outbreak by prevalent symptoms, population demographics and so forth and act accordingly. Moreover, Mallol argues, they can do it faster.

“I know of a product that (could map) all of the (Dengue) cases that happened in 2014, which took them about one year,” he says. “We do it in one minute or less.”

AIME plans to roll out their platform to other countries and regions. With visits to The Clinton Global Initiative and Harvard University under its belt, the co-founder says governments are more receptive to the start-up.

“Dengue is just the start,” he adds. “We will create a device to diagnose tuberculosis and malaria. We will create another software to diagnose diabetic retinopathy (a disease which can lead to blindness).” There’s also an idea “to link blood banks all over the world.”

It’s clear Mallol and AIME have no limits to their ambitions. Though for now, Dengue remains the focus.

“My invention will change the future,” he says. “I know this because it’s already changing the future in Malaysia.”