人工智能检测到的青光眼比研究中的医生更精确,有可能改进早期诊断。 An AI detected glaucoma more accurately than doctors in a study, potentially improving early diagnosis.

人工智能系统在检测青光眼方面优于人类专家,正确识别了88%至90%的病例的病情,而医生的病情为79%至81%,据美国眼科学会2024年会议提交的一项研究。 An AI system outperformed human experts in detecting glaucoma, correctly identifying the condition in 88% to 90% of cases compared to 79% to 81% for doctors, according to a study presented at the American Academy of Ophthalmology’s 2024 meeting. 人工智能使用垂直杯盘比分析了 6,300 多名参与者，这是液体压力导致眼睛损伤的关键迹象，尽管只有 11% 的眼睛出现青光眼迹象，但表现良好——反映了现实世界的筛查率。 The AI analyzed over 6,300 participants using the vertical cup-disc ratio, a key sign of eye damage from fluid pressure, and performed well even though only 11% of eyes showed glaucoma signs—reflecting real-world screening rates. 研究人员说,大赦国际可以改善获得早期检测的机会,特别是在服务不足的地区,如果与其他风险因素相结合,可以更加准确。 Researchers say AI could improve access to early detection, especially in underserved areas, and may become more accurate when combined with other risk factors. 调查结果是初步的,等待同行审议。 Findings are preliminary and await peer review.