AI system interprets eye scans as accurately as top specialistsBMJ 2018; 362 doi: https://doi.org/10.1136/bmj.k3484 (Published 13 August 2018) Cite this as: BMJ 2018;362:k3484
- Jacqui Wise
An artificial intelligence (AI) system recommended the correct referral decision for 53 eye diseases with 94% accuracy—matching the performance of top eye experts—according to new research published in Nature Medicine.1
The technology, which has been developed by researchers at Moorfields Eye Hospital NHS Foundation Trust, DeepMind Health, and University College London, needs to go through clinical trials and gain regulatory approval before it can be routinely used in hospitals. It is hoped, however, that the AI system could speed up the referral process for patients who urgently need treatment for diseases such as age related macular degeneration and diabetic retinopathy.
Optical coherence tomography (OCT) scans are used to help diagnose eye conditions. These are 3D images which provide a detailed map of the back of the eye but they are difficult to read and need …