实时使用深度学习准确识别房角镜下的小梁网。
Accurate Identification of the Trabecular Meshwork under Gonioscopic View in Real Time Using Deep Learning.
机构信息
Gavin Herbert Eye Institute, Department of Ophthalmology, University of California, Irvine, California; Department of Biomedical Engineering, University of California, Irvine, California.
Department of Computer Science, University of California, Irvine, California.
出版信息
Ophthalmol Glaucoma. 2022 Jul-Aug;5(4):402-412. doi: 10.1016/j.ogla.2021.11.003. Epub 2021 Nov 16.
PURPOSE
Accurate identification of iridocorneal structures on gonioscopy is difficult to master, and errors can lead to grave surgical complications. This study aimed to develop and train convolutional neural networks (CNNs) to accurately identify the trabecular meshwork (TM) in gonioscopic videos in real time for eventual clinical integrations.
DESIGN
Cross-sectional study.
PARTICIPANTS
Adult patients with open angle were identified in academic glaucoma clinics in both Taipei, Taiwan, and Irvine, California.
METHODS
Neural Encoder-Decoder CNNs (U-nets) were trained to predict a curve marking the TM using an expert-annotated data set of 378 gonioscopy images. The model was trained and evaluated with stratified cross-validation grouped by patients to ensure uncorrelated training and testing sets, as well as on a separate test set and 3 intraoperative gonioscopic videos of ab interno trabeculotomy with Trabectome (totaling 90 seconds long, 30 frames per second). We also evaluated our model's performance by comparing its accuracy against ophthalmologists.
MAIN OUTCOME MEASURES
Successful development of real-time-capable CNNs that are accurate in predicting and marking the TM's position in video frames of gonioscopic views. Models were evaluated in comparison with human expert annotations of static images and video data.
RESULTS
The best CNN model produced test set predictions with a median deviation of 0.8% of the video frame's height (15.25 μm) from the human experts' annotations. This error is less than the average vertical height of the TM. The worst test frame prediction of this model had an average deviation of 4% of the frame height (76.28 μm), which is still considered a successful prediction. When challenged with unseen images, the CNN model scored greater than 2 standard deviations above the mean performance of the surveyed general ophthalmologists.
CONCLUSIONS
Our CNN model can identify the TM in gonioscopy videos in real time with remarkable accuracy, allowing it to be used in connection with a video camera intraoperatively. This model can have applications in surgical training, automated screenings, and intraoperative guidance. The dataset developed in this study is one of the first publicly available gonioscopy image banks (https://lin.hs.uci.edu/research), which may encourage future investigations in this topic.
目的
准确识别房角镜下的虹膜角膜结构较难掌握，且错误可能导致严重的手术并发症。本研究旨在开发和训练卷积神经网络（CNN），以便实时准确地识别房角镜视频中的小梁网（TM），最终实现临床应用。
设计
横断面研究。
参与者
在台湾台北和加利福尼亚尔湾的学术青光眼诊所中，确定了患有开角型青光眼的成年患者。
方法
使用专家标记的 378 张房角镜图像数据集，对神经编码器-解码器 CNN（U 形网络）进行训练，以预测标记 TM 的曲线。通过对患者进行分层交叉验证来训练和评估模型，以确保训练和测试集无相关性，同时还使用独立的测试集和 3 段 Trabectome 内路小梁切开术中的房角镜视频（总时长 90 秒，每秒 30 帧）进行评估。我们还通过比较其准确性来评估模型的性能，将其与眼科医生进行比较。
主要观察指标
成功开发出实时 CNN，该 CNN 能够准确预测和标记房角镜视野视频帧中 TM 的位置。将模型与静态图像和视频数据的人类专家注释进行比较进行评估。
结果
最佳 CNN 模型对测试集的预测结果为，与人类专家注释相比，视频帧高度的偏差中位数为 0.8%（15.25 μm）。此误差小于 TM 的平均垂直高度。该模型预测的最差测试帧的平均偏差为帧高度的 4%（76.28 μm），仍被认为是成功的预测。当遇到未见图像时，CNN 模型的得分高于调查的普通眼科医生平均表现的 2 个标准差以上。
结论
我们的 CNN 模型可以实时准确地识别房角镜视频中的 TM，可用于术中连接视频摄像头。该模型可应用于手术培训、自动筛查和术中指导。本研究开发的数据集是首批公开发布的房角镜图像库之一（https://lin.hs.uci.edu/research），这可能会鼓励该领域的未来研究。
Zotero 插件