Automatic segmentation of cerebral infarcts in follow-up computed tomography images with convolutional neural networks

Sales Barros, R. et al. (2020) Automatic segmentation of cerebral infarcts in follow-up computed tomography images with convolutional neural networks. Journal of NeuroInterventional Surgery, 12(9), pp. 848-852. (doi: 10.1136/neurintsurg-2019-015471) (PMID:31871069)

[img]
Preview
Text
204048.pdf - Published Version
Available under License Creative Commons Attribution Non-commercial.

995kB

Abstract

Background and purpose: Infarct volume is a valuable outcome measure in treatment trials of acute ischemic stroke and is strongly associated with functional outcome. Its manual volumetric assessment is, however, too demanding to be implemented in clinical practice. Objective: To assess the value of convolutional neural networks (CNNs) in the automatic segmentation of infarct volume in follow-up CT images in a large population of patients with acute ischemic stroke. Materials and methods: We included CT images of 1026 patients from a large pooling of patients with acute ischemic stroke. A reference standard for the infarct segmentation was generated by manual delineation. We introduce three CNN models for the segmentation of subtle, intermediate, and severe hypodense lesions. The fully automated infarct segmentation was defined as the combination of the results of these three CNNs. The results of the three-CNNs approach were compared with the results from a single CNN approach and with the reference standard segmentations. Results: The median infarct volume was 48 mL (IQR 15–125 mL). Comparison between the volumes of the three-CNNs approach and manually delineated infarct volumes showed excellent agreement, with an intraclass correlation coefficient (ICC) of 0.88. Even better agreement was found for severe and intermediate hypodense infarcts, with ICCs of 0.98 and 0.93, respectively. Although the number of patients used for training in the single CNN approach was much larger, the accuracy of the three-CNNs approach strongly outperformed the single CNN approach, which had an ICC of 0.34. Conclusion: Convolutional neural networks are valuable and accurate in the quantitative assessment of infarct volumes, for both subtle and severe hypodense infarcts in follow-up CT images. Our proposed three-CNNs approach strongly outperforms a more straightforward single CNN approach.

Item Type:Articles
Additional Information:This study was funded by ITEA3 (grant number: 10004).
Status:Published
Refereed:Yes
Glasgow Author(s) Enlighten ID:Muir, Professor Keith
Authors: Sales Barros, R., Tolhuisen, M. L., Boers, A. M.M., Jansen, I., Ponomareva, E., Dippel, D. W.J., van der Lugt, A., van Oostenbrugge, R. J., van Zwam, W. H., Berkhemer, O. A., Goyal, M., Demchuk, A. M., Menon, B. K., Mitchell, P., Hill, M. D., Jovin, T. G., Davalos, A., Campbell, B. C.V., Saver, J. L., Roos, Y. B.W.E.M., Muir, K. W., White, P., Bracard, S., Guillemin, F., Delgado Olabarriaga, S., Majoie, C. B.L.M., and Marquering, H. A.
College/School:College of Medical Veterinary and Life Sciences > School of Psychology & Neuroscience
Journal Name:Journal of NeuroInterventional Surgery
Publisher:BMJ Publishing Group
ISSN:1759-8478
ISSN (Online):1759-8486
Published Online:23 December 2019
Copyright Holders:Copyright © 2019 The Authors
First Published:First published in Journal of NeuroInterventional Surgery 12(9): 848-852
Publisher Policy:Reproduced under a Creative Commons licence

University Staff: Request a correction | Enlighten Editors: Update this record