Artificial intelligence in ophthalmological practice: when ideal meets reality

BMJ Open Ophthalmology

article

Ludwig M Heindl¹  Senmao Li¹  Daniel S W Ting³  Pearse A Keane⁵ 

[1]Department of Ophthalmology , University of Cologne

[2]Department of Ophthalmology , The First Affiliated Hospital of Jinan University

[3]Singapore National Eye Center , Duke-NUS Medical School

[4]Ophthalmology and Visual Sciences Department , Duke-NUS Medical School

[5]Medical Retina , Moorfields Eye Hospital NHS Foundation Trust

关键词: diagnostic tests/investigation;    imaging;    public health;    telemedicine;    retina;   

DOI  :  10.1136/bmjophth-2022-001129

学科分类：工业工程学

来源: BMJ Publishing Group

【 摘 要 】

‘It was the best of times; it was the worst oftimes’. Such words never get old in any era.Similarly, in ophthalmology, the same thingis happening. Artificial intelligence (AI)using machine learning and deep learningin ophthalmology have created incrediblechemistry,1 2 with the US Food and DrugAdministration (FDA) approving AI-baseddiagnostic technology for autonomousdiabetic retinopathy screening.3 A humancentred AI study, on diabetic retinopathyscreening, conducted in Thailand with theGoogle team, highlighted many unexpectedreal-world problems and questions. Furtherwork is required to determine formal costbenefit analyses, specific workflows for liveimplementation in diverse healthcare settingsand solutions for real-world challenges suchas lack of internet connectivity or electronichealth records.

【 授权许可】

CC BY-NC|CC BY|CC BY-NC-ND