AI good for human health and that of global data science

Hoping to encourage interdisciplinary collaboration around Big Data over the next 50 to 100 years, Boston University is preparing to build a 17-story architectural marvel.

The school’s formidable news division has been covering the development in a series of articles. The latest looks at BU’s work with AI in healthcare.

For the piece, writer Art Jahnke spoke with scholars in medicine, pharmacology, global health, neurosurgery and other disciplines—not least data science.

For example, he describes the work of a data scientist who’s working with a systems engineer to help identify patients at risk of heart disease or diabetes. Their project will involve training algorithms on health data from EHRs as well as from wearable, implantable and home-based networked diagnostic devices.

“Eventually, we hope to move from predictions to prescriptions,” data scientist Yannis Paschalidis, PhD, tells the reporter. “We have some initial results for diabetes and hypertension. The goal is to make recommendations available in electronic health records as guidance for the care provider.”

Jahnke also introduces a professor of global health who’s working with a colleague in computer science to better understand Western influences on dietary trends in Kenya. The project has them analyzing millions of Instagram images posted by people living in the East African country.

Once they train their algorithm to recognize African foods, they’ll work to figure out “how ‘green’ or how ‘greasy’ the Kenyan diet is, and if it varies in urban and rural areas.”

Read the whole thing: