Artificial Intelligence in Stroke Risk Prediction

Uses

Benefits

Evidence quality

Where it comes from

Highlighted claims

• The ABSTRACT dataset includes 118,736 patients from NHS general practice and hospital records in Devon and Cornwall. — Opinions of the UK general public and stroke survivors in using artificial intelligence and ‘opt-out’ models of consent in medical research: a qualitative study
• The AI models are intended to predict stroke risk at 1, 3, 5, and 10 years using multiple clinical data types. — Opinions of the UK general public and stroke survivors in using artificial intelligence and ‘opt-out’ models of consent in medical research: a qualitative study
• All focus group and interview participants, and 60 of 61 survey respondents, considered AI training for stroke risk prediction appropriate. — Opinions of the UK general public and stroke survivors in using artificial intelligence and ‘opt-out’ models of consent in medical research: a qualitative study
• Support for AI use in stroke risk prediction was nearly universal among study participants. — Opinions of the UK general public and stroke survivors in using artificial intelligence and ‘opt-out’ models of consent in medical research: a qualitative study
• The study examined public acceptability of using NHS patient data to train AI models that predict future stroke risk. — Opinions of the UK general public and stroke survivors in using artificial intelligence and ‘opt-out’ models of consent in medical research: a qualitative study