深度学习算法在光学相干断层扫描中检测新生血管性年龄相关性黄斑变性特征:跨种族数据验证
该研究评估一个在韩国数据集上表现优异的深度学习模型,在外部美国数据集上识别新生血管性年龄相关性黄斑变性(nAMD)的泛化能力。外部验证中,个体水平AUC为0.952(95%CI 0.942–0.962),AUPRC为0.891(95%CI 0.875–0.908)。显著性图显示模型关注黄斑区与nAMD相关的病理特征,10名视网膜专家调查确认了这一点。
要点
- 深度学习模型在韩国数据开发后,在美国外部数据验证中AUC达0.952、AUPRC达0.891。
- 显著性图显示模型在正常OCT关注黄斑中心凹,在nAMD OCT关注相应病理特征。
- 10名视网膜专家的调查确认模型聚焦区域与nAMD病变一致。
核心结论
- 深度学习算法在韩国人群中识别nAMD表现优异,并在种族不同的美国人群中泛化良好。
- 模型正确聚焦黄斑区内差异以提取与nAMD相关的特征。
研究设计与样本
- 模型开发数据集为来自韩国的12 247张OCT扫描。
- 外部验证数据集为来自美国华盛顿的91 509张OCT扫描。
- 两个数据集均纳入正常眼或nAMD眼,内部测试后将算法送至美国华盛顿大学进行外部验证。
主要发现
- 外部验证中,个体水平AUC为0.952(95%CI 0.942至0.962)。
- 外部验证中,个体水平AUPRC为0.891(95%CI 0.875至0.908)。
- 显著性图显示正常OCT扫描中黄斑中心凹是主要关注区域。
- 显著性图显示nAMD OCT扫描中模型关注区域为适当的病理特征。
关键数值与推荐
- 外部验证AUC为0.952(95%CI 0.942至0.962),AUPRC为0.891(95%CI 0.875至0.908)。
- 10名视网膜专家的调查确认模型关注区域与nAMD病变一致。
临床怎么用
- 该深度学习算法可用于跨种族OCT数据中辅助识别nAMD,但需进一步临床验证。
- 模型显著性图可提供可解释性,帮助医生理解模型关注区域。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
BACKGROUND: The ability of deep learning (DL) algorithms to identify eyes with neovascular age-related macular degeneration (nAMD) from optical coherence tomography (OCT) scans has been previously established. We herewith evaluate the ability of a DL model, showing excellent performance on a Korean data set, to generalse onto an American data set despite ethnic differences. In addition, expert graders were surveyed to verify if the DL model was appropriately identifying lesions indicative of nAMD on the OCT scans. METHODS: Model development data set-12 247 OCT scans from South Korea; external validation data set-91 509 OCT scans from Washington, USA. In both data sets, normal eyes or eyes with nAMD were included. After internal testing, the algorithm was sent to the University of Washington, USA, for external validation. Area under the receiver operating characteristic curve (AUC) and precision-recall curve (AUPRC) were calculated. For model explanation, saliency maps were generated using Guided GradCAM. RESULTS: On external validation, AUC and AUPRC remained high at 0.952 (95% CI 0.942 to 0.962) and 0.891 (95% CI 0.875 to 0.908) at the individual level. Saliency maps showed that in normal OCT scans, the fovea was the main area of interest; in nAMD OCT scans, the appropriate pathological features were areas of model interest. Survey of 10 retina specialists confirmed this. CONCLUSION: Our DL algorithm exhibited high performance for nAMD identification in a Korean population, and generalised well to an ethnically distinct, American population. The model correctly focused on the differences within the macular area to extract features associated with nAMD.
引用
- Rim TH, Lee AY, Ting DS, et al. Detection of features associated with neovascular age-related macular degeneration in ethnically distinct data sets by an optical coherence tomography: trained deep learning algorithm. Br J Ophthalmol. 2020. PMID: 32907811. doi:10.1136/bjophthalmol-2020-316984
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。