Mila > News > Mila researchers Sasha Luccioni and Charles Onu selected as Harvard CRCS 2021 Rising Stars for their work in the field of AI for social good

24 Mar 2021

Mila researchers Sasha Luccioni and Charles Onu selected as Harvard CRCS 2021 Rising Stars for their work in the field of AI for social good

/

AI for social good projects are increasingly gaining momentum, and Mila researchers whose work falls within this category are getting noticed. PhD student Charles C Onu (McGill) and postdoctoral researcher Sasha Luccioni (UdeM) have been selected as part of the 2021 class of Rising Stars by the Harvard Center for Research on Computation and Society (CRCS) for their outstanding AI for social good endeavors.

Continued advances in machine learning and artificial intelligence present new opportunities to develop cutting-edge tools and solutions that can help tackle some of the world’s most pressing issues. Starting March 30, the Harvard CRCS Rising Stars Speaker Series will feature presentations from 16 PhD and postdoctoral candidates who were nominated by AI experts for their exemplary work in the fields of public health, conservation, fairness, and technology and society.

During the first talk in the series, Onu will be presenting the Ubenwa mobile app, a screening tool that can identify respiratory distress in newborns by analyzing the amplitude and frequency of cry patterns and estimate the risk of birth asphyxia. The following session on conservation will include a talk by Luccioni on “This Climate Does Not Exist,” an upcoming interactive website that will serve as an educational tool and help the public visualize climate change in a unique and interactive way.

Leveraging AI to detect respiratory distress in newborns

Onu, who holds a Vanier Canada Graduate Scholarship, focuses his research at the intersection of AI, healthcare and biomedical engineering. He is the Founder and AI Research Lead at Ubenwa Health, a startup whose mission is to make clinical diagnostic tools more widely accessible.

Growing up in Nigeria, of all the socioeconomic challenges around me, the deficiencies in healthcare stood out the most. I’ve lost relatives due to simple inadequacies. It was always a question of how I can apply my science and engineering education to improve healthcare and my path led me to AI. – Charles C Onu, PhD student at Mila and McGill, and AI Research Lead at Ubenwa Health.

The multidisciplinary team of researchers at Ubenwa has developed an easy-to-use mobile app that uses AI to analyze the sound of newborn babies’ cries to detect early signs of perinatal asphyxia, a condition in which a baby’s brain is deprived of oxygen before, during, or immediately following birth. According to the World Health Organization, perinatal asphyxia accounts for 4 million deaths annually, making it one of the top causes of mortality and life-long disabilities in infants worldwide.

Ubenwa, which means “cry of a baby” in the Igbo language of Nigeria, could have a major impact in mitigating this global health problem, especially in resource-limited regions or countries. This non-invasive tool allows clinicians to detect the risk of asphyxia early, treat it, and potentially save millions of lives.

Ubenwa Health is also exploring how their app could identify several other conditions that affect newborns’ central and autonomic nervous systems. “Our goals for the year include completing our ongoing clinical studies in Canada and Nigeria, releasing a beta version of the Ubenwa app, and securing additional funding to grow the team and prepare for clinical trials,” added Onu.

Imagining the real impacts of climate change, one address at a time, thanks to AI

Luccioni, who holds an IVADO postdoctoral scholarship and is a 2020 National Geographic Explorer, earning her the AI for Earth Innovation Grant, is leading a research project at the nexus of machine learning and climate change. Luccioni and her team are currently developing “This Climate Does Not Exist,” an immersive website focused on user-experience that can generate realistic images of the future impacts of climate change.

By training complex neural networks known as GANs on before-and-after images of extreme climate events, these GANs can now generate vivid renderings of the effects of floods, smog and wildfire on people’s own homes and the places they hold dear, whether it’s their house, neighbourhood, or favourite travel destination.

When asked what prompted her to focus her efforts on climate change, Luccioni said: “I wanted to tuck my daughters in at night and tell them that mommy’s doing her best to protect their future. Climate change is such a huge challenge that will affect so many aspects of our current and future society, so for me, it’s really important to do my part to tackle it.”

With the website set to launch in June 2021, the goal of the project is to educate the public in order to compel them to put climate change at the forefront of their everyday choices. According to Luccioni, the key to bringing about collective and individual action is through education and awareness.

“Once people realize that climate change isn’t just some distant, abstract notion, they will make the necessary changes in their lives. That’s our hope with this project, to use AI as a way to pique people’s curiosity and inspire them to take action.”


For more information:

Visit the Harvard CRCS Rising Stars information page to learn more and to register for the speaker series.

Conferences on Public Health and Conservation:

Tuesday, March 30, 2021: Click here to register for the talk on Public Health featuring Charles Onu.

Thursday, April 8, 2021: Click here to register for the talk on Conservation with Sasha Luccioni.

All events will take place from 12:00-1:30 pm ET.