多模态机器学习联合视野与视盘周围环形OCT扫描检测青光眼性视神经病变
本研究开发并验证了多模态人工智能算法FusionNet,利用视野报告中的模式偏差概率图和环形视盘周围OCT扫描检测青光眼性视神经病变。FusionNet在主验证集、内部测试集和外部测试集中诊断性能均优于单模态VFNet和OCTNet,与两位青光眼专家的AUC相当。
要点
- FusionNet在主验证集AUC为0.950(95%CI 0.931-0.968),优于VFNet、OCTNet及两位青光眼专家。
- 内部测试集AUC:FusionNet 0.917,VFNet 0.854,OCTNet 0.811;外部测试集:0.873、0.772、0.785。
- 多模态融合视野与OCT数据可提升青光眼性视神经病变的检测效能。
核心结论
- FusionNet使用配对的视野和OCT数据开发,在检测青光眼性视神经病变方面优于VFNet和OCTNet。
- 多模态机器学习模型在检测青光眼性视神经病变方面具有价值。
研究设计与样本
- 数据分为训练集1567对(HFA和DRI-OCT)、主验证集441对(HFA和DRI-OCT)、内部测试集255对(HFA和Cirrus OCT)和外部测试集200对(HFA和Spectralis)。
主要发现
- 主验证集中FusionNet的AUC为0.950(95%CI 0.931-0.968),优于VFNet(AUC 0.868,95%CI 0.834-0.902)、OCTNet(AUC 0.809,95%CI 0.768-0.850)和两位青光眼专家(专家1 AUC 0.882,95%CI 0.847-0.917;专家2 AUC 0.883,95%CI 0.849-0.918)。
- 内部测试集AUC:FusionNet 0.917,VFNet 0.854,OCTNet 0.811;外部测试集AUC:FusionNet 0.873,VFNet 0.772,OCTNet 0.785。
- 内部和外部测试集中,两位青光眼专家与FusionNet之间无显著差异,但内部测试集中专家2(AUC 0.858,95%CI 0.805-0.912)除外。
关键数值与推荐
- 青光眼性视神经病变定义为视网膜神经纤维层变薄伴相应视野缺损。
临床怎么用
- FusionNet可作为辅助工具,联合视野和OCT数据提升青光眼性视神经病变的检测准确性。
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
PURPOSE: To develop and validate a multimodal artificial intelligence algorithm, FusionNet, using the pattern deviation probability plots from visual field (VF) reports and circular peripapillary OCT scans to detect glaucomatous optic neuropathy (GON). DESIGN: Cross-sectional study. SUBJECTS: Two thousand four hundred sixty-three pairs of VF and OCT images from 1083 patients. METHODS: FusionNet based on bimodal input of VF and OCT paired data was developed to detect GON. Visual field data were collected using the Humphrey Field Analyzer (HFA). OCT images were collected from 3 types of devices (DRI-OCT, Cirrus OCT, and Spectralis). Two thousand four hundred sixty-three pairs of VF and OCT images were divided into 4 datasets: 1567 for training (HFA and DRI-OCT), 441 for primary validation (HFA and DRI-OCT), 255 for the internal test (HFA and Cirrus OCT), and 200 for the external test set (HFA and Spectralis). GON was defined as retinal nerve fiber layer thinning with corresponding VF defects. MAIN OUTCOME MEASURES: Diagnostic performance of FusionNet compared with that of VFNet (with VF data as input) and OCTNet (with OCT data as input). RESULTS: FusionNet achieved an area under the receiver operating characteristic curve (AUC) of 0.950 (0.931-0.968) and outperformed VFNet (AUC, 0.868 [95% confidence interval (CI), 0.834-0.902]), OCTNet (AUC, 0.809 [95% CI, 0.768-0.850]), and 2 glaucoma specialists (glaucoma specialist 1: AUC, 0.882 [95% CI, 0.847-0.917]; glaucoma specialist 2: AUC, 0.883 [95% CI, 0.849-0.918]) in the primary validation set. In the internal and external test sets, the performances of FusionNet were also superior to VFNet and OCTNet (FusionNet vs VFNet vs OCTNet: internal test set 0.917 vs 0.854 vs 0.811; external test set 0.873 vs 0.772 vs 0.785). No significant difference was found between the 2 glaucoma specialists and FusionNet in the internal and external test sets, except for glaucoma specialist 2 (AUC, 0.858 [95% CI, 0.805-0.912]) in the internal test set. CONCLUSIONS: FusionNet, developed using paired VF and OCT data, demonstrated superior performance to both VFNet and OCTNet in detecting GON, suggesting that multimodal machine learning models are valuable in detecting GON.
引用
- Xiong J, Li F, Song D, et al. Multimodal Machine Learning Using Visual Fields and Peripapillary Circular OCT Scans in Detection of Glaucomatous Optic Neuropathy. Ophthalmology. 2021. PMID: 34339778. doi:10.1016/j.ophtha.2021.07.032
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。