A Deep Learning Approach for Accurate Discrimination Between Optic Disc Drusen and Papilledema on Fundus Photographs.

IF 2 4区医学 Q3 CLINICAL NEUROLOGY

Journal of Neuro-Ophthalmology Pub Date : 2024-12-01 Epub Date: 2024-08-02 DOI:10.1097/WNO.0000000000002223

Kanchalika Sathianvichitr, Raymond P Najjar, Tang Zhiqun, J Alexander Fraser, Christine W L Yau, Michael J A Girard, Fiona Costello, Mung Y Lin, Wolf A Lagrèze, Catherine Vignal-Clermont, Clare L Fraser, Steffen Hamann, Nancy J Newman, Valérie Biousse, Dan Milea

{"title":"A Deep Learning Approach for Accurate Discrimination Between Optic Disc Drusen and Papilledema on Fundus Photographs.","authors":"Kanchalika Sathianvichitr, Raymond P Najjar, Tang Zhiqun, J Alexander Fraser, Christine W L Yau, Michael J A Girard, Fiona Costello, Mung Y Lin, Wolf A Lagrèze, Catherine Vignal-Clermont, Clare L Fraser, Steffen Hamann, Nancy J Newman, Valérie Biousse, Dan Milea","doi":"10.1097/WNO.0000000000002223","DOIUrl":null,"url":null,"abstract":"Background: Optic disc drusen (ODD) represent an important differential diagnosis of papilledema caused by intracranial hypertension, but their distinction may be difficult in clinical practice. The aim of this study was to train, validate, and test a dedicated deep learning system (DLS) for binary classification of ODD vs papilledema (including various subgroups within each category), on conventional mydriatic digital ocular fundus photographs collected in a large international multiethnic population.Methods: This retrospective study included 4,508 color fundus images in 2,180 patients from 30 neuro-ophthalmology centers (19 countries) participating in the Brain and Optic Nerve Study with Artificial Intelligence (BONSAI) Group. For training and internal validation, we used 857 ODD images and 3,230 papilledema images, in 1,959 patients. External testing was performed on an independent data set (221 patients), including 207 images with ODD (96 visible and 111 buried), provided by 3 centers of the Optic Disc Drusen Studies Consortium, and 214 images of papilledema (92 mild-to-moderate and 122 severe) from a previously validated study.Results: The DLS could accurately distinguish between all ODD and papilledema (all severities included): area under the receiver operating characteristic curve (AUC) 0.97 (95% confidence interval [CI], 0.96-0.98), accuracy 90.5% (95% CI, 88.0%-92.9%), sensitivity 86.0% (95% CI, 82.1%-90.1%), and specificity 94.9% (95% CI, 92.3%-97.6%). The performance of the DLS remained high for discrimination of buried ODD from mild-to-moderate papilledema: AUC 0.93 (95% CI, 0.90-0.96), accuracy 84.2% (95% CI, 80.2%-88.6%), sensitivity 78.4% (95% CI, 72.2%-84.7%), and specificity 91.3% (95% CI, 87.0%-96.4%).Conclusions: A dedicated DLS can accurately distinguish between ODD and papilledema caused by intracranial hypertension, even when considering buried ODD vs mild-to-moderate papilledema.","PeriodicalId":16485,"journal":{"name":"Journal of Neuro-Ophthalmology","volume":" ","pages":"454-461"},"PeriodicalIF":2.0000,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Neuro-Ophthalmology","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1097/WNO.0000000000002223","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2024/8/2 0:00:00","PubModel":"Epub","JCR":"Q3","JCRName":"CLINICAL NEUROLOGY","Score":null,"Total":0}

引用次数: 0

Abstract

Background: Optic disc drusen (ODD) represent an important differential diagnosis of papilledema caused by intracranial hypertension, but their distinction may be difficult in clinical practice. The aim of this study was to train, validate, and test a dedicated deep learning system (DLS) for binary classification of ODD vs papilledema (including various subgroups within each category), on conventional mydriatic digital ocular fundus photographs collected in a large international multiethnic population.

Methods: This retrospective study included 4,508 color fundus images in 2,180 patients from 30 neuro-ophthalmology centers (19 countries) participating in the Brain and Optic Nerve Study with Artificial Intelligence (BONSAI) Group. For training and internal validation, we used 857 ODD images and 3,230 papilledema images, in 1,959 patients. External testing was performed on an independent data set (221 patients), including 207 images with ODD (96 visible and 111 buried), provided by 3 centers of the Optic Disc Drusen Studies Consortium, and 214 images of papilledema (92 mild-to-moderate and 122 severe) from a previously validated study.

Results: The DLS could accurately distinguish between all ODD and papilledema (all severities included): area under the receiver operating characteristic curve (AUC) 0.97 (95% confidence interval [CI], 0.96-0.98), accuracy 90.5% (95% CI, 88.0%-92.9%), sensitivity 86.0% (95% CI, 82.1%-90.1%), and specificity 94.9% (95% CI, 92.3%-97.6%). The performance of the DLS remained high for discrimination of buried ODD from mild-to-moderate papilledema: AUC 0.93 (95% CI, 0.90-0.96), accuracy 84.2% (95% CI, 80.2%-88.6%), sensitivity 78.4% (95% CI, 72.2%-84.7%), and specificity 91.3% (95% CI, 87.0%-96.4%).

Conclusions: A dedicated DLS can accurately distinguish between ODD and papilledema caused by intracranial hypertension, even when considering buried ODD vs mild-to-moderate papilledema.

查看原文本刊更多论文

在眼底照片上准确区分视盘色素沉着和乳头水肿的深度学习方法。

背景：视盘色素（ODD）是颅内高压引起的乳头水肿的一个重要鉴别诊断，但在临床实践中可能难以区分。本研究的目的是训练、验证和测试一个专用的深度学习系统（DLS），以在一个大型国际多种族人群中收集到的常规瞳孔数字眼底照片上对 ODD 与乳头水肿（包括每个类别中的各种亚组）进行二元分类：这项回顾性研究包括来自 30 个神经眼科中心（19 个国家）的 2,180 名患者的 4,508 张彩色眼底图像，这些中心参加了脑与视神经人工智能研究（BONSAI）小组。在训练和内部验证中，我们使用了 1,959 名患者的 857 张 ODD 图像和 3,230 张乳头水肿图像。外部测试是在一个独立的数据集（221 名患者）上进行的，其中包括由视盘日晕研究联盟的 3 个中心提供的 207 张视盘日晕图像（96 张可见，111 张被掩盖），以及之前验证过的一项研究提供的 214 张乳头水肿图像（92 张轻度至中度，122 张重度）：DLS 能准确区分所有 ODD 和乳头水肿（包括所有严重程度）：接收者工作特征曲线下面积 (AUC) 为 0.97（95% 置信区间 [CI]，0.96-0.98），准确率为 90.5%（95% CI，88.0%-92.9%），灵敏度为 86.0%（95% CI，82.1%-90.1%），特异性为 94.9%（95% CI，92.3%-97.6%）。在区分埋藏性乳头状水肿和轻度至中度乳头状水肿方面，DLS 的性能仍然很高：AUC为0.93（95% CI，0.90-0.96），准确率为84.2%（95% CI，80.2%-88.6%），灵敏度为78.4%（95% CI，72.2%-84.7%），特异性为91.3%（95% CI，87.0%-96.4%）：即使考虑到埋藏性 ODD 与轻度至中度乳头水肿，专用的 DLS 也能准确区分 ODD 与颅内高压引起的乳头水肿。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Journal of Neuro-Ophthalmology 医学-临床神经学

CiteScore

2.80

自引率

13.80%

发文量

593

审稿时长

6-12 weeks

期刊介绍： The Journal of Neuro-Ophthalmology (JNO) is the official journal of the North American Neuro-Ophthalmology Society (NANOS). It is a quarterly, peer-reviewed journal that publishes original and commissioned articles related to neuro-ophthalmology.