Automated lumen segmentation using multi-frame convolutional neural networks in intravascular ultrasound datasets
Carregando...
Citações na Scopus
8
Tipo de produção
article
Data de publicação
2020
Título da Revista
ISSN da Revista
Título do Volume
Editora
OXFORD UNIV PRESS
Autores
ZIEMER, Paulo G. P.
BULANT, Carlos A.
I, Jose Orlando
TALOU, Gonzalo D. Maso
ALVAREZ, Luis A. Mansilla
GARCIA-GARCIA, Hector M.
BLANCO, Pablo J.
Citação
EUROPEAN HEART JOURNAL - DIGITAL HEALTH, v.1, n.1, p.75-82, 2020
Resumo
AimsAssessment of minimum lumen areas in intravascular ultrasound (IVUS) pullbacks is time-consuming and demands adequately trained personnel. In this work, we introduce a novel and fully automated pipeline to segment the lumen boundary in IVUS datasets.Methods and resultsFirst, an automated gating is applied to select end-diastolic frames and bypass saw-tooth artefacts. Second, within a machine learning (ML) environment, we automatically segment the lumen boundary using a multi-frame (MF) convolutional neural network (MFCNN). Finally, we use the theory of Gaussian processes (GPs) to regress the final lumen boundary. The dataset consisted of 85 IVUS pullbacks (52 patients). The dataset was partitioned at the pullback-level using 73 pullbacks for training (20 586 frames), 6 pullbacks for validation (1692 frames), and 6 for testing (1692 frames). The degree of overlapping, between the ground truth and ML contours, median (interquartile range, IQR) systematically increased from 0.896 (0.874-0.933) for MF1 to 0.925 (0.911-0.948) for MF11. The median (IQR) of the distance error was also reduced from 3.83 (2.94-4.98)% for MF1 to 3.02 (2.25-3.95)% for MF11-GP. The corresponding median (IQR) in the lumen area error remained between 5.49 (2.50-10.50)% for MF1 and 5.12 (2.15-9.00)% for MF11-GP. The dispersion in the relative distance and area errors consistently decreased as we increased the number of frames, and also when the GP regressor was coupled to the MFCNN output.ConclusionThese results demonstrate that the proposed ML approach is suitable to effectively segment the lumen boundary in IVUS scans, reducing the burden of costly and time-consuming manual delineation.
Palavras-chave
IVUS, Segmentation, Gating, Neural networks, Gaussian process
Referências
- Balakrishna C, 2018, Arxiv, DOI [arXiv:1806.07554, DOI 10.48550/ARXIV.1806.07554,ARXIV]
- Cui HF, 2020, INT J NUMER METH BIO, V36, DOI 10.1002/cnm.3348
- de Winter SA, 2004, CATHETER CARDIO INTE, V61, P84, DOI 10.1002/ccd.10693
- Talou GDM, 2015, IEEE T BIO-MED ENG, V62, P2867, DOI 10.1109/TBME.2015.2449232
- Mintz GS, 2001, J AM COLL CARDIOL, V37, P1478, DOI 10.1016/S0735-1097(01)01175-5
- Neumann FJ, 2019, EUR HEART J, V40, P79, DOI [10.1093/eurheartj/ehy855, 10.15829/1560-4071-2019-8-151-226]
- Rasmussen C.E., 2006, Gaussian Processes for Machine Learning, P248
- Ronneberger O, 2015, LECT NOTES COMPUT SC, V9351, P234, DOI 10.1007/978-3-319-24574-4_28
- Taha AA, 2015, BMC MED IMAGING, V15, DOI 10.1186/s12880-015-0068-x
- Yang J, 2018, Arxiv, DOI arXiv:1806.03583