一种用于在光谱域光学相干断层扫描中检测和分类脉络膜新生血管的机器学习算法的准确性

Accuracy of a Machine-Learning Algorithm for Detecting and Classifying Choroidal Neovascularization on Spectral-Domain Optical Coherence Tomography.

作者信息

Maunz Andreas, Benmansour Fethallah, Li Yvonna, Albrecht Thomas, Zhang Yan-Ping, Arcadu Filippo, Zheng Yalin, Madhusudhan Savita, Sahni Jayashree

机构信息

Pharma Research and Early Development, Roche Innovation Center, F. Hoffmann-La Roche Ltd., 4070 Basel, Switzerland.

Department of Eye and Vision Science, University of Liverpool, Liverpool L7 8XP, UK.

出版信息

J Pers Med. 2021 Jun 8;11(6):524. doi: 10.3390/jpm11060524.

BACKGROUND

To evaluate the performance of a machine-learning (ML) algorithm to detect and classify choroidal neovascularization (CNV), secondary to age-related macular degeneration (AMD) on spectral-domain optical coherence tomography (SD-OCT) images.

METHODS

Baseline fluorescein angiography (FA) and SD-OCT images from 1037 treatment-naive study eyes and 531 fellow eyes, without advanced AMD from the phase 3 HARBOR trial (NCT00891735), were used to develop, train, and cross-validate an ML pipeline combining deep-learning-based segmentation of SD-OCT B-scans and CNV classification, based on features derived from the segmentations, in a five-fold setting. FA classification of the CNV phenotypes from HARBOR was used for generating the ground truth for model development. SD-OCT scans from the phase 2 AVENUE trial (NCT02484690) were used to externally validate the ML model.

RESULTS

The ML algorithm discriminated CNV absence from CNV presence, with a very high accuracy (area under the receiver operating characteristic [AUROC] = 0.99), and classified occult versus predominantly classic CNV types, per FA assessment, with a high accuracy (AUROC = 0.91) on HARBOR SD-OCT images. Minimally classic CNV was discriminated with significantly lower performance. Occult and predominantly classic CNV types could be discriminated with AUROC = 0.88 on baseline SD-OCT images of 165 study eyes, with CNV from AVENUE.

CONCLUSIONS

Our ML model was able to detect CNV presence and CNV subtypes on SD-OCT images with high accuracy in patients with neovascular AMD.

背景

评估一种机器学习（ML）算法在光谱域光学相干断层扫描（SD-OCT）图像上检测和分类年龄相关性黄斑变性（AMD）继发的脉络膜新生血管（CNV）的性能。

方法

来自3期HARBOR试验（NCT00891735）的1037只未经治疗的研究眼和531只对侧眼的基线荧光素血管造影（FA）和SD-OCT图像，这些眼睛无晚期AMD，用于开发、训练和交叉验证一个ML流程，该流程结合基于深度学习的SD-OCT B扫描分割和基于分割衍生特征的CNV分类，采用五折设置。来自HARBOR的CNV表型的FA分类用于为模型开发生成真值。来自2期AVENUE试验（NCT02484690）的SD-OCT扫描用于对ML模型进行外部验证。

结果

ML算法能够以非常高的准确率（受试者操作特征曲线下面积[AUROC]=0.99）区分有无CNV，并根据FA评估，在HARBOR SD-OCT图像上以高准确率（AUROC=0.91）对隐匿性与主要为典型性CNV类型进行分类。对最小经典型CNV的区分性能显著较低。在165只研究眼的基线SD-OCT图像上，隐匿性和主要为经典型CNV类型的区分AUROC=0.88，这些研究眼来自AVENUE试验且存在CNV。