Joint Learning of Vessel Segmentation and Artery/Vein Classification with Post-processing

Liangzhi Li, Manisha Verma, Yuta Nakashima, Ryo Kawasaki, Hajime Nagahara
; Proceedings of the Third Conference on Medical Imaging with Deep Learning, PMLR 121:440-453, 2020.

Abstract

Retinal imaging serves as a valuable tool for diagnosis of various diseases. However, reading retinal images is a difficult and time-consuming task even for experienced specialists. The fundamental step towards automated retinal image analysis is vessel segmentation and artery/vein classification, which provide various information on potential disorders. To improve the performance of the existing automated methods for retinal image analysis, we propose a two-step vessel classification. We adopt a UNet-based model, SeqNet, to accurately segment vessels from the background and make prediction on the vessel type. Our model does segmentation and classification sequentially, which alleviates the problem of label distribution bias and facilitates training. To further refine classification results, we post-process them considering the structural information among vessels to propagate highly confident prediction to surrounding vessels. Our experiments show that our method improves AUC to 0.98 for segmentation and the accuracy to 0.92 in classification over DRIVE dataset.

Cite this Paper


BibTeX
@InProceedings{pmlr-v121-li20a, title = {Joint Learning of Vessel Segmentation and Artery/Vein Classification with Post-processing}, author = {Li, Liangzhi and Verma, Manisha and Nakashima, Yuta and Kawasaki, Ryo and Nagahara, Hajime}, pages = {440--453}, year = {2020}, editor = {Tal Arbel and Ismail Ben Ayed and Marleen de Bruijne and Maxime Descoteaux and Herve Lombaert and Christopher Pal}, volume = {121}, series = {Proceedings of Machine Learning Research}, address = {Montreal, QC, Canada}, month = {06--08 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v121/li20a/li20a.pdf}, url = {http://proceedings.mlr.press/v121/li20a.html}, abstract = {Retinal imaging serves as a valuable tool for diagnosis of various diseases. However, reading retinal images is a difficult and time-consuming task even for experienced specialists. The fundamental step towards automated retinal image analysis is vessel segmentation and artery/vein classification, which provide various information on potential disorders. To improve the performance of the existing automated methods for retinal image analysis, we propose a two-step vessel classification. We adopt a UNet-based model, SeqNet, to accurately segment vessels from the background and make prediction on the vessel type. Our model does segmentation and classification sequentially, which alleviates the problem of label distribution bias and facilitates training. To further refine classification results, we post-process them considering the structural information among vessels to propagate highly confident prediction to surrounding vessels. Our experiments show that our method improves AUC to 0.98 for segmentation and the accuracy to 0.92 in classification over DRIVE dataset.} }
Endnote
%0 Conference Paper %T Joint Learning of Vessel Segmentation and Artery/Vein Classification with Post-processing %A Liangzhi Li %A Manisha Verma %A Yuta Nakashima %A Ryo Kawasaki %A Hajime Nagahara %B Proceedings of the Third Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2020 %E Tal Arbel %E Ismail Ben Ayed %E Marleen de Bruijne %E Maxime Descoteaux %E Herve Lombaert %E Christopher Pal %F pmlr-v121-li20a %I PMLR %J Proceedings of Machine Learning Research %P 440--453 %U http://proceedings.mlr.press %V 121 %W PMLR %X Retinal imaging serves as a valuable tool for diagnosis of various diseases. However, reading retinal images is a difficult and time-consuming task even for experienced specialists. The fundamental step towards automated retinal image analysis is vessel segmentation and artery/vein classification, which provide various information on potential disorders. To improve the performance of the existing automated methods for retinal image analysis, we propose a two-step vessel classification. We adopt a UNet-based model, SeqNet, to accurately segment vessels from the background and make prediction on the vessel type. Our model does segmentation and classification sequentially, which alleviates the problem of label distribution bias and facilitates training. To further refine classification results, we post-process them considering the structural information among vessels to propagate highly confident prediction to surrounding vessels. Our experiments show that our method improves AUC to 0.98 for segmentation and the accuracy to 0.92 in classification over DRIVE dataset.
APA
Li, L., Verma, M., Nakashima, Y., Kawasaki, R. & Nagahara, H.. (2020). Joint Learning of Vessel Segmentation and Artery/Vein Classification with Post-processing. Proceedings of the Third Conference on Medical Imaging with Deep Learning, in PMLR 121:440-453

Related Material