基于人工智能的光学相干断层扫描血管造影术对吸烟人群眼底血流密度变化的研究
Fundus blood flow density changes in the smoking population by artificial intelligence-based optical coherence tomography angiography.
作者信息
Zhang Ling-Yu, Li Qing-Jian, Zhou Qiang, Zhang Yu, Liu Yan, Wang Zhi-Liang, Zhang Pei
机构信息
Tianjin Key Laboratory of Retinal Functions and Diseases, Tianjin Branch of National Clinical Research Center for Ocular Disease, Eye Institute and School of Optometry, Tianjin Medical University Eye Hospital, Tianjin 300380, China.
Department of Ophthalmology, Huashan Hospital, Fudan University, Shanghai 200040, China.
出版信息
Int J Ophthalmol. 2025 Sep 18;18(9):1613-1618. doi: 10.18240/ijo.2025.09.01. eCollection 2025.
AIM
To determine whether chronic smoking affects fundus blood flow density using optical coherence tomography angiography (OCTA) based on artificial intelligence (AI).
METHODS
All participants underwent a comprehensive ophthalmological examination in this study. The subjects were categorized into two groups: control and smoker. Fundus data obtained through the novel OCTA device were compared.
RESULTS
Utilizing deep learning denoising techniques removed background noise and smoothed vessel surfaces. OCTA showed a significant decrease in fundus blood flow density after AI-based denoising on the right eyes of 36 smokers (36 males, average age 44.17±9.85y) and age- and sex-matched participants who never smoked. The thickness of the retina in both control and smoker groups failed to show any statistically significant differences. Smoking was associated with decreased blood flow density in the macula and the optic disk.
CONCLUSION
Utilizing AI-based denoising to improve the sensitivity of OCTA images can be highly beneficial.
目的
使用基于人工智能（AI）的光学相干断层扫描血管造影（OCTA）来确定长期吸烟是否会影响眼底血流密度。
方法
在本研究中，所有参与者均接受了全面的眼科检查。受试者被分为两组：对照组和吸烟者。对通过新型OCTA设备获得的眼底数据进行了比较。
结果
利用深度学习去噪技术去除了背景噪声并平滑了血管表面。OCTA显示，在36名吸烟者（36名男性，平均年龄44.17±9.85岁）以及年龄和性别匹配的从不吸烟者的右眼上，基于AI的去噪后眼底血流密度显著降低。对照组和吸烟者组的视网膜厚度均未显示出任何统计学上的显著差异。吸烟与黄斑和视盘处的血流密度降低有关。
结论
利用基于AI的去噪来提高OCTA图像的灵敏度可能非常有益。
Zotero 插件