Deep learning for calcium segmentation in intravascular ultrasound images
Standard
Deep learning for calcium segmentation in intravascular ultrasound images. / Bargsten, Lennart; Riedl, Katharina A.; Wissel, Tobias; Brunner, Fabian J.; Schaefers, Klaus; Grass, Michael; Blankenberg, Stefan; Seiffert, Moritz; Schlaefer, Alexander.
In: Current Directions in Biomedical Engineering, Vol. 7, No. 1, 20211123, 01.08.2021.Research output: SCORING: Contribution to journal › SCORING: Journal article › Research › peer-review
Harvard
APA
Vancouver
Bibtex
}
RIS
TY - JOUR
T1 - Deep learning for calcium segmentation in intravascular ultrasound images
AU - Bargsten, Lennart
AU - Riedl, Katharina A.
AU - Wissel, Tobias
AU - Brunner, Fabian J.
AU - Schaefers, Klaus
AU - Grass, Michael
AU - Blankenberg, Stefan
AU - Seiffert, Moritz
AU - Schlaefer, Alexander
N1 - Publisher Copyright: © 2021 by Walter de Gruyter Berlin/Boston.
PY - 2021/8/1
Y1 - 2021/8/1
N2 - Knowing the shape of vascular calcifications is crucial for appropriate planning and conductance of percutaneous coronary interventions. The clinical workflow can therefore benefit from automatic segmentation of calcified plaques in intravascular ultrasound (IVUS) images. To solve segmentation problems with convolutional neural networks (CNNs), large datasets are usually required. However, datasets are often rather small in the medical domain. Hence, developing and investigating methods for increasing CNN performance on small datasets can help on the way towards clinically relevant results. We compared two state-of-the-art CNN architectures for segmentation, U-Net and DeepLabV3, and investigated how incorporating auxiliary image data with vessel wall and lumen annotations improves the calcium segmentation performance by using these either for pretraining or multi-task training. DeepLabV3 outperforms U-Net with up to 6.3 % by means of the Dice coefficient and 36.5 % by means of the average Hausdorff distance. Using auxiliary data improves the segmentation performance in both cases, whereas the multi-task approach outperforms the pre-training approach. The improvements of the multi-task approach in contrast to not using auxiliary data at all is 5.7 % for the Dice coefficient and 42.9 % for the average Hausdorff distance. Automatic segmentation of calcified plaques in IVUS images is a demanding task due to their relatively small size compared to the image dimensions and due to visual ambiguities with other image structures. We showed that this problem can generally be tackled by CNNs. Furthermore, we were able to improve the performance by a multi-task learning approach with auxiliary segmentation data.
AB - Knowing the shape of vascular calcifications is crucial for appropriate planning and conductance of percutaneous coronary interventions. The clinical workflow can therefore benefit from automatic segmentation of calcified plaques in intravascular ultrasound (IVUS) images. To solve segmentation problems with convolutional neural networks (CNNs), large datasets are usually required. However, datasets are often rather small in the medical domain. Hence, developing and investigating methods for increasing CNN performance on small datasets can help on the way towards clinically relevant results. We compared two state-of-the-art CNN architectures for segmentation, U-Net and DeepLabV3, and investigated how incorporating auxiliary image data with vessel wall and lumen annotations improves the calcium segmentation performance by using these either for pretraining or multi-task training. DeepLabV3 outperforms U-Net with up to 6.3 % by means of the Dice coefficient and 36.5 % by means of the average Hausdorff distance. Using auxiliary data improves the segmentation performance in both cases, whereas the multi-task approach outperforms the pre-training approach. The improvements of the multi-task approach in contrast to not using auxiliary data at all is 5.7 % for the Dice coefficient and 42.9 % for the average Hausdorff distance. Automatic segmentation of calcified plaques in IVUS images is a demanding task due to their relatively small size compared to the image dimensions and due to visual ambiguities with other image structures. We showed that this problem can generally be tackled by CNNs. Furthermore, we were able to improve the performance by a multi-task learning approach with auxiliary segmentation data.
KW - Convolutional neural network
KW - Coronary artery
KW - Multi-task learning
KW - Small dataset
KW - Vessel
UR - http://www.scopus.com/inward/record.url?scp=85114431339&partnerID=8YFLogxK
U2 - 10.1515/cdbme-2021-1021
DO - 10.1515/cdbme-2021-1021
M3 - SCORING: Journal article
AN - SCOPUS:85114431339
VL - 7
JO - Current Directions in Biomedical Engineering
JF - Current Directions in Biomedical Engineering
SN - 2364-5504
IS - 1
M1 - 20211123
ER -