人工智能和眼科临床注册表
Luke Tran1, Himal Kandel1, Daliya Sari1
1From the Faculty of Medicine and Health, Save Sight Institute, The University of Sydney, (L.T., H.K., D.S., C.H.C., S.L.W.) Sydney, New South Wales, Australia.
American journal of ophthalmology
|August 7, 2024
概括
人工智能 (AI) 对医疗保健有希望,但AI模型需要广泛的数据. 本综述探讨了眼科临床注册表中的AI应用,发现先进AI的使用有限,需要标准化验证.
科学领域:
- 眼科医生 眼科 眼科
- 人工智能的人工智能
- 医疗信息学 医疗信息学
背景情况:
- 人工智能 (AI) 为增加临床需求和有限的医疗资源提供了解决方案.
- 人工智能模型需要大型,具有代表性的数据集来准确的临床预测.
- 眼科临床注册表是训练人工智能的真实世界数据的宝贵来源.
研究的目的:
- 审查人工智能 (AI) 在眼科临床注册数据中的当前应用.
- 识别用于注册表数据的AI算法类型和数据输入的趋势.
- 评估眼科注册表中人工智能的当前状态和未来潜力.
主要方法:
- 在2024年7月在EMBASE,Medline,PubMed,Scopus和Web of Science进行了系统的文献搜索.
- 包括应用AI到眼科临床注册数据的初级研究文章.
- 分析了AI算法类型,数据输入,输出和验证方法.
主要成果:
- 确定了23篇主要研究文章,重点关注14个眼科注册表.
- 玻璃眼和新血管与年龄相关的黄斑变性是研究的最常见的疾病.
- 监督的传统机器学习模型占主导地位 (85%),深度学习或NLP的使用有限;在验证中观察到显著的异质性.
结论:
- 人工智能在眼科临床注册表中的应用还处于新生阶段,深度学习的利用有限.
- 数据可访问性差以及缺乏标准化验证方法阻碍了人工智能开发.
- 未来的进步需要标准化的方法和更多的领域专家参与临床部署AI.
相关概念视频
01:25Glaucoma: Overview
Glaucoma is an eye condition characterized by increased intraocular pressure that damages the retina and optic nerve, leading to irreversible blindness if left untreated. The human eye has various components, including the cornea, iris, pupil, lens, and optic nerve. Aqueous humor is secreted by the epithelium of the ciliary body in the posterior chamber and flows through the trabecular meshwork and canal of Schlemm, maintaining normal intraocular pressure. The trabecular meshwork and the canal...
01:23Ophthalmic Drug Delivery Systems
Ophthalmic drug delivery faces major limitations due to poor absorption across the corneal membrane. This process is primarily driven by diffusion and is influenced by two main factors: the physicochemical properties of the drug and tear drainage. Most ophthalmic drugs, such as pilocarpine, epinephrine, atropine, and local anesthetics, are weak bases. They are typically formulated at an acidic pH to enhance chemical stability. However, this leads to high ionization, reducing their ability to...