Semi-supervised learning of cardiac MRI using image registration

Autor: Garcia-Cabrera, Carles, Curran, Kathleen M., O'Connor, Noel E., McGuinness, Kevin
Jazyk: angličtina
Rok vydání: 2021
Předmět:
Zdroj: Garcia-Cabrera, Carles, Curran, Kathleen M. ORCID: 0000-0003-0095-9337 , O'Connor, Noel E. ORCID: 0000-0002-4033-9135 and McGuinness, Kevin ORCID: 0000-0003-1336-6477 (2021) Semi-supervised learning of cardiac MRI using image registration. In: Irish Machine Vision and Image Processing Conference (IMVIP), 1-3 Sept 2021, Dublin, Ireland. ISBN 978-0-9934207-6-4
Popis: In this work, we propose a method to aid the 2-D segmentation of short-axis cardiac MRI. In particular, the deformation fields obtained during the registration are used to propagate the labels to all time frames, resulting in a weakly supervised segmentation approach that benefits from the features in unlabelled volumes along with the annotated data. Experimental results over the M\&Ms datasets show that the addition of the synthetically obtained labels to the original dataset yields promising results in the performance and improves the capability of the network to generalise to scanners from different vendors.
Databáze: OpenAIRE