Cargando…

One-shot Retinal Artery and Vein Segmentation via Cross-modality Pretraining

PURPOSE: To perform one-shot retinal artery and vein segmentation with cross-modality artery-vein (AV) soft-label pretraining. DESIGN: Cross-sectional study. SUBJECTS: The study included 6479 color fundus photography (CFP) and arterial-venous fundus fluorescein angiography (FFA) pairs from 1964 part...

Descripción completa

Detalles Bibliográficos
Autores principales: Shi, Danli, He, Shuang, Yang, Jiancheng, Zheng, Yingfeng, He, Mingguang
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Elsevier 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10585631/
https://www.ncbi.nlm.nih.gov/pubmed/37868792
http://dx.doi.org/10.1016/j.xops.2023.100363
_version_ 1785122993971658752
author Shi, Danli
He, Shuang
Yang, Jiancheng
Zheng, Yingfeng
He, Mingguang
author_facet Shi, Danli
He, Shuang
Yang, Jiancheng
Zheng, Yingfeng
He, Mingguang
author_sort Shi, Danli
collection PubMed
description PURPOSE: To perform one-shot retinal artery and vein segmentation with cross-modality artery-vein (AV) soft-label pretraining. DESIGN: Cross-sectional study. SUBJECTS: The study included 6479 color fundus photography (CFP) and arterial-venous fundus fluorescein angiography (FFA) pairs from 1964 participants for pretraining and 6 AV segmentation data sets with various image sources, including RITE, HRF, LES-AV, AV-WIDE, PortableAV, and DRSplusAV for one-shot finetuning and testing. METHODS: We structurally matched the arterial and venous phase of FFA with CFP, the AV soft labels were automatically generated by utilizing the fluorescein intensity difference of the arterial and venous-phase FFA images, and the soft labels were then used to train a generative adversarial network to learn to generate AV soft segmentations using CFP images as input. We then finetuned the pretrained model to perform AV segmentation using only one image from each of the AV segmentation data sets and test on the remainder. To investigate the effect and reliability of one-shot finetuning, we conducted experiments without finetuning and by finetuning the pretrained model on an iteratively different single image for each data set under the same experimental setting and tested the models on the remaining images. MAIN OUTCOME MEASURES: The AV segmentation was assessed by area under the receiver operating characteristic curve (AUC), accuracy, Dice score, sensitivity, and specificity. RESULTS: After the FFA-AV soft label pretraining, our method required only one exemplar image from each camera or modality and achieved similar performance with full-data training, with AUC ranging from 0.901 to 0.971, accuracy from 0.959 to 0.980, Dice score from 0.585 to 0.773, sensitivity from 0.574 to 0.763, and specificity from 0.981 to 0.991. Compared with no finetuning, the segmentation performance improved after one-shot finetuning. When finetuned on different images in each data set, the standard deviation of the segmentation results across models ranged from 0.001 to 0.10. CONCLUSIONS: This study presents the first one-shot approach to retinal artery and vein segmentation. The proposed labeling method is time-saving and efficient, demonstrating a promising direction for retinal-vessel segmentation and enabling the potential for widespread application. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.
format Online
Article
Text
id pubmed-10585631
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher Elsevier
record_format MEDLINE/PubMed
spelling pubmed-105856312023-10-20 One-shot Retinal Artery and Vein Segmentation via Cross-modality Pretraining Shi, Danli He, Shuang Yang, Jiancheng Zheng, Yingfeng He, Mingguang Ophthalmol Sci Original Articles PURPOSE: To perform one-shot retinal artery and vein segmentation with cross-modality artery-vein (AV) soft-label pretraining. DESIGN: Cross-sectional study. SUBJECTS: The study included 6479 color fundus photography (CFP) and arterial-venous fundus fluorescein angiography (FFA) pairs from 1964 participants for pretraining and 6 AV segmentation data sets with various image sources, including RITE, HRF, LES-AV, AV-WIDE, PortableAV, and DRSplusAV for one-shot finetuning and testing. METHODS: We structurally matched the arterial and venous phase of FFA with CFP, the AV soft labels were automatically generated by utilizing the fluorescein intensity difference of the arterial and venous-phase FFA images, and the soft labels were then used to train a generative adversarial network to learn to generate AV soft segmentations using CFP images as input. We then finetuned the pretrained model to perform AV segmentation using only one image from each of the AV segmentation data sets and test on the remainder. To investigate the effect and reliability of one-shot finetuning, we conducted experiments without finetuning and by finetuning the pretrained model on an iteratively different single image for each data set under the same experimental setting and tested the models on the remaining images. MAIN OUTCOME MEASURES: The AV segmentation was assessed by area under the receiver operating characteristic curve (AUC), accuracy, Dice score, sensitivity, and specificity. RESULTS: After the FFA-AV soft label pretraining, our method required only one exemplar image from each camera or modality and achieved similar performance with full-data training, with AUC ranging from 0.901 to 0.971, accuracy from 0.959 to 0.980, Dice score from 0.585 to 0.773, sensitivity from 0.574 to 0.763, and specificity from 0.981 to 0.991. Compared with no finetuning, the segmentation performance improved after one-shot finetuning. When finetuned on different images in each data set, the standard deviation of the segmentation results across models ranged from 0.001 to 0.10. CONCLUSIONS: This study presents the first one-shot approach to retinal artery and vein segmentation. The proposed labeling method is time-saving and efficient, demonstrating a promising direction for retinal-vessel segmentation and enabling the potential for widespread application. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article. Elsevier 2023-07-06 /pmc/articles/PMC10585631/ /pubmed/37868792 http://dx.doi.org/10.1016/j.xops.2023.100363 Text en © 2023 by the American Academy of Ophthalmology. https://creativecommons.org/licenses/by-nc-nd/4.0/This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
spellingShingle Original Articles
Shi, Danli
He, Shuang
Yang, Jiancheng
Zheng, Yingfeng
He, Mingguang
One-shot Retinal Artery and Vein Segmentation via Cross-modality Pretraining
title One-shot Retinal Artery and Vein Segmentation via Cross-modality Pretraining
title_full One-shot Retinal Artery and Vein Segmentation via Cross-modality Pretraining
title_fullStr One-shot Retinal Artery and Vein Segmentation via Cross-modality Pretraining
title_full_unstemmed One-shot Retinal Artery and Vein Segmentation via Cross-modality Pretraining
title_short One-shot Retinal Artery and Vein Segmentation via Cross-modality Pretraining
title_sort one-shot retinal artery and vein segmentation via cross-modality pretraining
topic Original Articles
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10585631/
https://www.ncbi.nlm.nih.gov/pubmed/37868792
http://dx.doi.org/10.1016/j.xops.2023.100363
work_keys_str_mv AT shidanli oneshotretinalarteryandveinsegmentationviacrossmodalitypretraining
AT heshuang oneshotretinalarteryandveinsegmentationviacrossmodalitypretraining
AT yangjiancheng oneshotretinalarteryandveinsegmentationviacrossmodalitypretraining
AT zhengyingfeng oneshotretinalarteryandveinsegmentationviacrossmodalitypretraining
AT hemingguang oneshotretinalarteryandveinsegmentationviacrossmodalitypretraining