Researchers at the University at Buffalo wrote a meta-review, published in NPJ Digital Medicine. They screened about 5,000 peer-reviewed studies and selected 60 that look at AI and wearable devices for people with Type 2 diabetes and prediabetes. Many wearables, like continuous glucose monitors (CGMs), give frequent glucose readings. AI models can use these data to recognise patterns and predict glucose changes.
The review finds positives and limits. AI-enhanced wearables can predict glucose changes one to two hours in advance and help people keep steadier glucose control. They can also sort large data and reduce some clinical work. For people with prediabetes, early use of wearables plus AI could support lifestyle changes and possibly delay diabetes. But the research is uneven, some models are hard to understand, and device cost and access remain problems.
Difficult words
- meta-review — a study that reviews many other studies
- wearable — a small device a person can wearwearables
- continuous glucose monitor — a device that measures blood sugar continuouslycontinuous glucose monitors
- recognise — see or find a pattern in data
- predict — say what will happen before it occurs
- prediabetes — a health condition before type two diabetes
- steady — not changing much over timesteadier
- access — ability to get something or use it
Tip: hover, focus or tap highlighted words in the article to see quick definitions while you read or listen.
Discussion questions
- Would you use a wearable to monitor your health? Why or why not?
- What is one positive effect of AI-enhanced wearables mentioned in the article?
- Which problem about the devices or research worries you most? Why?
Related articles
Forest loss in tropics raises local heat and deaths
A study using satellite data found that tropical deforestation from 2001–2020 exposed 345 million people to local warming and likely caused about 28,000 heat-related deaths per year, mainly in Africa, Southeast Asia and Latin America.
