Doing good with machine learning and AI

Share on twitter
Share on facebook
Share on linkedin

Times of catastrophe or crisis often bring humans together to help each other with good deeds. It’s no surprise then that humanitarian groups have been using artificial intelligence and machine learning to predict, understand, and manage how we respond to disasters, epidemics, and other emergencies.

We know that AI and ML are great additions to any business toolkit and, used properly, they can also help those on the front lines of tragedy.

How humanitarians are leveraging the power of AI

Here are a few examples of how humanitarians have leveraged the power of artificial intelligence and machine learning to assist those in need.

Delivering emergency medical aid

Médecins Sans Frontières, or Doctors Without Borders, is an international aid organization that focuses on delivering emergency medical aid quickly to where it’s needed. The nurses, doctors, and other professionals affiliated with MSF often need to react and deploy quickly to a medical hotspot. They need to make sure that they’re coordinating with local authorities to sensure that the right services are delivered at the right time in the right place.

To help support this decision-marking, the organization developed the Reaction Assessment Collaboration Hub (REACH) to map and analyze often rapidly changing situations. REACH “combines institutional data with crowd-sourced information, drawing on a vast number of sources.” Combining all these sources and making sense of them is a perfect use for machine learning, and it allows MSF to “collate unstructured data into one secure portal, helping to provide the best analysis and plan the optimal response.”

Medical professionals on the ground can access this information via a chatbot that gives them the information they need in a timely and efficient manner.

Responding to natural disasters

The response to natural disasters like earthquakes, floods, and typhoons are another area where AI and ML are often leveraged. Following the aftermath of Hurricane Katrina, which struck the Southeastern US in 2005, scientists and rescuers used Unmanned Aerial Vehicles (UAVs) that can cover large distances, to search for survivors. Using ML-based prediction algorithms, rescuers were often able to determine where survivors were likely go. The UAVs were often autonomous and operated quickly to direct searchers to the right place to find and rescue survivors.

Helping to stop world hunger

Hunger is a persistent problem across the globe, with estimates that more than 820 million people suffer from lack of proper nutrition. In order to help the hungry, researchers first needed determine what areas have the greatest need. They did this by using AI to analyze satellite imagery to find areas of poverty.

Humanitarians and technologists are collaborating to use AI and ML in a variety of ways to help alleviate this entrenched problem. Some projects include using AI to improve agricultural yields, helping farmers plant crops at the right time by combining historical yield data, satellite imagery, and climate information into ML-based predictive models. Soil composition analysis using the data generated by a variety of sensors also helps increase crop yields.

Predicting health outcomes & lowering readmissions

Health outcomes are another area where AI and ML are used to great benefit to patients, physicians, and researchers.

If you’re a data scientist, or otherwise data science-adjacent, the Center for Medicare and Medicaid Services (a US government agency) is running a competition (worth US$1 million to the winner) to devise a program that uses AI tools to predict health outcomes for patients admitted to hospitals or other skilled nursing facilities.

Hospital and skilled nursing readmissions are both costly and dangerous. With this competition, they’re hoping to harness AI to predict health outcomes and lower the number of readmissions.

Conclusion

ML and AI aren’t just about the bottom line. They have real potential to improve human lives in a variety of domains if their power is brought to bear.

If you’d like to learn more about getting started with a machine learning project—whether for your business or for humanitarian reasons—check out our Human’s Guide to Machine Learning Projects. It walks you through the critical first steps of a new project to help you succeed.

Looking to drive real business impact with AI?

Sometimes the most difficult thing is simply knowing where to start. Identifying impactful use cases is one of the most cited roadblocks for organizations seeking to leverage AI.

Learn from these 50 use cases across all industries.

Additional Reading

Chris Doty

Chris Doty

Chris Doty is RapidMiner's Content Marketing Manager and has worked on projects for companies like Google, NatGeo TV, the US Chamber of Commerce, and Virgin Pulse. In a past life he was an academic, and has a PhD in linguistics.