Digital@UNGA 2025 spotlights digital solutions, innovations and investments for humanity as the United Nations marks 80 years.

MIT researchers developed an interactive, AI-based system that enables users to rapidly annotate areas of interest in new biomedical imaging datasets, without training a machine-learning model in advance. As the user uploads new images, the number of interactions needed to accurately segment the image drops, eventually to zero, enabling rapid annotation of the entire dataset.

Stroke patients evaluated using telemedicine (telestroke) have higher odds of receiving essential treatment, yet it takes them significantly longer to be treated — potentially limiting the benefits, a study finds. Telestroke patients had 44% lower odds of being successfully treated within 60 minutes of getting to the hospital.