Could Artificial Intelligence Help Us Predict the Next Epidemic?

The medical startup AIME has successfully combined public health data and artificial intelligence to come up with a method of predicting the outbreak of epidemics before they even happen.

Author Julian Furtkamp:

Translation Julian Furtkamp, 06.27.17

The Artificial Intelligence and Medical Epidemiology platform (otherwise known as AIME Inc) aims to aid in the prevention of diseases by using artificial intelligence and machine learning to predict the outbreak of epidemics. Set up in Malaysia by the epidemiologist Dr Dhesi Raja and Singularity University graduate Rainier Mallol, their algorithm analyses a large set of data from different sources to determine the site of new outbreaks of dengue fever.

Dengue is an illness that a huge 40% of the population are at risk of contracting, and with a huge 390 million cases each year, according to statistics from the WHO. The virus’s reach already stretches all the way from Central and South America, through Africa and Asia right over to Australia. And as temperatures rise, the number of cases is increasing.

Predicting Dengue Outbreaks With A.I.

The system developed by AIME analyses not only public health data, but other data from other sources, such as weather, wind speed, previous outbreaks and a location’s proximity to large bodies of water – anything which might influence the behaviour of the mosquitoes that carry the disease. They also look at things like population density in the area, peoples’ health records and their income level.

The project started in Malaysia and recently spread further afield, with the AIME team collaborating with NGOs and local governments in Brazil before the Olympic Games in 2016, to carry out a pilot programme and develop preventative strategies to halt the spread of zika and dengue.

Their system claims to be able to provide the exact geo location and date of the next dengue outbreak, up to three months in advance, and can recommend anti-dengue measures for the area within a 400-metre radius – including genetically modified mosquitoes and so-called fogging.

For more examples of how state-of-the-art technologies are helping tackle mosquito-borne diseases around the world, read about how SolarMal is battling mosquitoes with the power of the sun, and the acoustic sensors measuring mozzies’ wingbeats in Indonesia. And get the lowdown on the link between poverty and disease in our comprehensive Knowledge article.

This article is a translation by Marisa Pettit of the original which first appeared on RESET’s German-language platform.

BreatheDX: Diagnosing Lung Diseases With a Smartphone

BreatheDX is a low-cost kit that can be used together with a smartphone to diagnose lung diseases in areas where health care services are lacking.

Wazi Vision: Bringing Affordable Eye Care to East African Children

A Ugandan start-up uses an app to diagnose eye problems among school going children, and provides them cheap glasses made of recycled materials. Wazi Vision wants to bring eye care to those who can’t afford traditional methods.

SolarMal: How To Fight Malaria Mozzies with the Power Of the Sun

In Africa, a child dies from malaria every minute while in the Americas, the alarming spread of the Zika virus is endangering the health of millions of people. Now scientists have developed a mosquito trap that could help reduce the number of infections.

Portable Community Healthcare That’s Inside the Box

Loaded with medical tools and technologies, the RxBox enables healthcare workers to provide a broader scope of healthcare to isolated, remote or disadvantaged communities in the Philippines.

Using a Startup Model to Bring Healthcare to Low-income Earners in Kenya

Penda Health, a healthcare social enterprise in Nairobi, Kenya, is bringing affordable medical services to low-income families via out-of-the-box thinking and mobile banking.

HIV Tests on Your Mobile: Could a Smartphone Save Your Life?

Scientists at Columbia University have developed an affordable and innovative device that plugs into smartphones and can test for both HIV and syphilis, delivering a diagnosis in as little as 15 minutes.