Automated Detection of Posterior Vitreous Detachment on OCT Using Computer Vision and Deep Learning Algorithms

To develop automated algorithms for the detection of posterior vitreous detachment (PVD) using OCT imaging. Evaluation of a diagnostic test or technology. Overall, 42 385 consecutive OCT images (865 volumetric OCT scans) obtained with Heidelberg Spectralis from 865 eyes from 464 patients at an acade...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	Ophthalmology science (Online) Jg. 3; H. 2; S. 100254
Hauptverfasser:	Li, Alexa L., Feng, Moira, Wang, Zixi, Baxter, Sally L., Huang, Lingling, Arnett, Justin, Bartsch, Dirk-Uwe G., Kuo, David E., Saseendrakumar, Bharanidharan Radha, Guo, Joy, Nudleman, Eric
Format:	Journal Article
Sprache:	Englisch
Veröffentlicht:	Netherlands Elsevier Inc 01.06.2023 Elsevier
Schlagworte:	Artificial Intelligence and Big Data Automated detection Deep Learning OCT Ophthalmology Posterior vitreous detachment ViT OCT CNN Automated detection AI DL AUROC ILM Deep Learning Posterior vitreous detachment PVD internal limiting membrane convolutional neural network area under the receiver operator characteristic curve artificial intelligence vision transformers AI, artificial intelligence ViT, vision transformers PVD, posterior vitreous detachment AUROC, area under the receiver operator characteristic curve DL, deep learning CNN, convolutional neural network ILM, internal limiting membrane
ISSN:	2666-9145, 2666-9145
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Beschreibung
Zusammenfassung:	To develop automated algorithms for the detection of posterior vitreous detachment (PVD) using OCT imaging. Evaluation of a diagnostic test or technology. Overall, 42 385 consecutive OCT images (865 volumetric OCT scans) obtained with Heidelberg Spectralis from 865 eyes from 464 patients at an academic retina clinic between October 2020 and December 2021 were retrospectively reviewed. We developed a customized computer vision algorithm based on image filtering and edge detection to detect the posterior vitreous cortex for the determination of PVD status. A second deep learning (DL) image classification model based on convolutional neural networks and ResNet-50 architecture was also trained to identify PVD status from OCT images. The training dataset consisted of 674 OCT volume scans (33 026 OCT images), while the validation testing set consisted of 73 OCT volume scans (3577 OCT images). Overall, 118 OCT volume scans (5782 OCT images) were used as a separate external testing dataset. Accuracy, sensitivity, specificity, F1-scores, and area under the receiver operator characteristic curves (AUROCs) were measured to assess the performance of the automated algorithms. Both the customized computer vision algorithm and DL model results were largely in agreement with the PVD status labeled by trained graders. The DL approach achieved an accuracy of 90.7% and an F1-score of 0.932 with a sensitivity of 100% and a specificity of 74.5% for PVD detection from an OCT volume scan. The AUROC was 89% at the image level and 96% at the volume level for the DL model. The customized computer vision algorithm attained an accuracy of 89.5% and an F1-score of 0.912 with a sensitivity of 91.9% and a specificity of 86.1% on the same task. Both the computer vision algorithm and the DL model applied on OCT imaging enabled reliable detection of PVD status, demonstrating the potential for OCT-based automated PVD status classification to assist with vitreoretinal surgical planning. Proprietary or commercial disclosure may be found after the references.
Bibliographie:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 A.L.L. and M.F. contributed equally to this work.
ISSN:	2666-9145 2666-9145
DOI:	10.1016/j.xops.2022.100254