Options
Infratentorial lesions in multiple sclerosis patients: intra- and inter-rater variability in comparison to a fully automated segmentation using 3D convolutional neural networks
Publikationstyp
Journal Article
Date Issued
2022-04
Sprache
English
Journal
Volume
32
Issue
4
Start Page
2798
End Page
2809
Citation
European Radiology 32 (4): 2798-2809 (2022-04)
Publisher DOI
Scopus ID
Publisher
Springer
Objective: Automated quantification of infratentorial multiple sclerosis lesions on magnetic resonance imaging is clinically relevant but challenging. To overcome some of these problems, we propose a fully automated lesion segmentation algorithm using 3D convolutional neural networks (CNNs). Methods: The CNN was trained on a FLAIR image alone or on FLAIR and T1-weighted images from 1809 patients acquired on 156 different scanners. An additional training using an extra class for infratentorial lesions was implemented. Three experienced raters manually annotated three datasets from 123 MS patients from different scanners. Results: The inter-rater sensitivity (SEN) was 80% for supratentorial lesions but only 62% for infratentorial lesions. There was no statistically significant difference between the inter-rater SEN and the SEN of the CNN with respect to the raters. For supratentorial lesions, the CNN featured an intra-rater intra-scanner SEN of 0.97 (R1 = 0.90, R2 = 0.84) and for infratentorial lesion a SEN of 0.93 (R1 = 0.61, R2 = 0.73). Conclusion: The performance of the CNN improved significantly for infratentorial lesions when specifically trained on infratentorial lesions using a T1 image as an additional input and matches the detection performance of experienced raters. Furthermore, for infratentorial lesions the CNN was more robust against repeated scans than experienced raters. Key Points: • A 3D convolutional neural network was trained on MRI data from 1809 patients (156 different scanners) for the quantification of supratentorial and infratentorial multiple sclerosis lesions. • Inter-rater variability was higher for infratentorial lesions than for supratentorial lesions. The performance of the 3D convolutional neural network (CNN) improved significantly for infratentorial lesions when specifically trained on infratentorial lesions using a T1 image as an additional input. • The detection performance of the CNN matches the detection performance of experienced raters.
Subjects
Artificial intelligence
Infratentorial lesions
Intra- and inter-rater variability
Lesion segmentation
Multiple sclerosis
MLE@TUHH
DDC Class
600: Technik
610: Medizin
Funding Organisations
More Funding Information
This work was partially funded by Zentrales Innovationsprogramm Mittelstand (ZIM) (contract number ZF4268403TS9) and by Hamburgische Investitions- und Förderbank (IFB) (contract number 51084589).
Sven Schippling reports compensation for consulting, serving on scientific advisory boards, speaking, or other activities from Biogen, Celgene, Merck, Sanofi, and TEVA. Hagen H. Kitzler has received travel grants, speaker’s honoraria, financial research support, and consultancy fees from Bayer, Biogen Idec, Novartis, Siemens, and TEVA. He served on advisory boards for Biogen, Novartis, and Ixico. He received research grants from Novartis.
Sven Schippling reports compensation for consulting, serving on scientific advisory boards, speaking, or other activities from Biogen, Celgene, Merck, Sanofi, and TEVA. Hagen H. Kitzler has received travel grants, speaker’s honoraria, financial research support, and consultancy fees from Bayer, Biogen Idec, Novartis, Siemens, and TEVA. He served on advisory boards for Biogen, Novartis, and Ixico. He received research grants from Novartis.