基于深度学习的三维未分割光学相干断层扫描体积与视神经萎缩疾病的关联
这项横断面研究使用多个临床试验和转诊中心2008至2025年的数据,训练三维深度学习模型分析未分割的视神经乳头光学相干断层扫描(OCT)扫描,以区分青光眼、非动脉炎性前部缺血性视神经病变(NAION)、视神经炎和健康眼。全容积模型准确率达88.9%(宏观AUC-ROC 0.977,95%CI 0.974-0.979),视神经炎分类最具挑战性。作者认为该自动化方法可能支持诊断努力并指导临床管理。
要点
- 全容积模型对青光眼、NAION、视神经炎和健康眼的F1分数分别为0.94、0.87、0.78和0.91。
- 仅视乳头周围区域模型准确率85.9%(AUC-ROC 0.970,95%CI 0.967-0.972),仅视神经乳头模型87.0%(AUC-ROC 0.972,95%CI 0.970-0.975)。
- 视神经炎在轴索丢失严重或轻微时最易被误分类为NAION或健康眼。
核心结论
- 基于未分割OCT扫描的深度学习分析能可靠区分不同形式的视神经萎缩,提示存在细微的疾病特异性结构模式。
- 这种自动化方法可能支持诊断工作、指导视神经病变的临床管理,并补充标准化程度较低的影像模式和主观临床印象。
研究设计与样本
- 横断面研究,使用2008至2025年多个临床试验和转诊中心的数据,包括随机试验、纵向研究和转诊诊所。
- 共分析1382只眼的7014次Cirrus视神经乳头OCT扫描,其中青光眼113例、NAION 311例、视神经炎163例、健康对照715例。
- 训练了三个ResNet-3D-18模型,分别评估全OCT体积、仅视乳头周围区域和仅视神经乳头,采用5折分层交叉验证,使用相同数据划分以直接比较性能。
主要发现
- 平均(标准差)年龄为54.2(16.9)岁;男性733例(65%),女性402例(35%)。
- 全容积模型准确率88.9%(宏观AUC-ROC 0.977,95%CI 0.974-0.979),青光眼、NAION、视神经炎和健康眼的F1分数分别为0.94、0.87、0.78和0.91。
- 仅视乳头周围区域模型准确率85.9%(AUC-ROC 0.970,95%CI 0.967-0.972),仅视神经乳头模型准确率87.0%(AUC-ROC 0.972,95%CI 0.970-0.975),两者F1分数范围为0.71至0.94。
- 视神经炎分类挑战最大,当轴索丢失严重或轻微时被误分类为NAION或健康眼。
- 激活图揭示了视网膜中疾病特异性感兴趣区域,包括视网膜神经纤维层、神经节细胞层和视网膜色素上皮。
关键数值与推荐
- 全容积模型:准确率88.9%,宏观AUC-ROC 0.977(95%CI 0.974-0.979),青光眼F1 0.94、NAION F1 0.87、视神经炎F1 0.78、健康眼F1 0.91。
- 仅视乳头周围区域模型:准确率85.9%,AUC-ROC 0.970(95%CI 0.967-0.972)。
- 仅视神经乳头模型:准确率87.0%,AUC-ROC 0.972(95%CI 0.970-0.975)。
临床怎么用
- 原文未据此给出推荐。
局限性
- 原文未报告局限性。
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
IMPORTANCE: Accurate differentiation of optic nerve head (ONH) atrophy is vital for guiding diagnosis and treatment of conditions such as glaucoma, nonarteritic anterior ischemic optic neuropathy (NAION), and optic neuritis. Traditional 2-dimensional assessments may overlook subtle, volumetric changes. OBJECTIVE: To determine whether a 3-dimensional (3D) deep learning model trained on unsegmented ONH optical coherence tomography (OCT) scans can reliably distinguish optic atrophy in glaucoma, NAION, optic neuritis, and healthy eyes. DESIGN, SETTING, AND PARTICIPANTS: This cross-sectional study used data from multiple clinical trials and referral centers (2008-2025), including randomized trials, longitudinal studies, and referral clinics. Participants included patients with glaucoma, NAION, or optic neuritis and healthy control patients. EXPOSURES: Three ResNet-3D-18 models were trained using 5-fold stratified cross-validation. One assessed the full OCT volume, another focused only on the peripapillary region (PPR), and the third considered only the ONH. Identical data splits were used to allow direct performance comparison. MAIN OUTCOMES AND MEASURES: Classification accuracy, macro area under the receiver operating characteristic curve (AUC-ROC), precision, recall, and F1 scores, aggregated across all validation folds. Confusion matrices were generated to characterize misclassifications. RESULTS: A total of 7014 Cirrus ONH OCT scans from 1382 eyes of glaucoma (n = 113), NAION (n = 311), optic neuritis (n = 163), and healthy controls (n = 715) were analyzed. The mean (SD) age was 54.2 (16.9) years; there were 733 (65%) male patients and 402 (35%) female patients. The entire-volume model achieved 88.9% accuracy (macro AUC-ROC, 0.977; 95% CI, 0.974-0.979) and F1 scores of 0.94, 0.87, 0.78, and 0.91 for glaucoma, NAION, optic neuritis, and healthy eyes, respectively. The PPR-only model reached 85.9% accuracy (AUC-ROC, 0.970; 95% CI, 0.967-0.972), while the ONH-only model attained 87.0% accuracy (AUC-ROC, 0.972; 95% CI, 0.970-0.975). Both achieved F1 scores from 0.71 to 0.94. Optic neuritis presented the greatest classification challenge, misclassified as NAION or healthy when axonal loss was severe or minimal. Activation maps revealed disease-specific regions of interest in the retina, including the retinal nerve fiber layer, ganglion cell layer, and retinal pigment epithelium. CONCLUSIONS AND RELEVANCE: Deep learning-based analysis of unsegmented OCT scans reliably distinguished between different forms of optic nerve atrophy, suggesting subtle, disease-specific structural patterns. This automated approach may support diagnostic efforts, guide clinical management of optic neuropathies, and complement less standardized imaging modalities and subjective clinical impressions.
引用
- Szanto D, Wang JK, Woods B, et al. Optic Nerve Atrophy Conditions Associated With 3D Unsegmented Optical Coherence Tomography Volumes Using Deep Learning. JAMA Ophthalmol. 2025. PMID: 40839281. doi:10.1001/jamaophthalmol.2025.2766
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。