IRUS Total

Deep generative model-based quality control for cardiac MRI segmentation

File Description SizeFormat 
2006.13379v1.pdfAccepted version1.04 MBAdobe PDFView/Open
Title: Deep generative model-based quality control for cardiac MRI segmentation
Authors: Wang, S
Tarroni, G
Qin, C
Mo, Y
Dai, C
Chen, C
Glocker, B
Guo, Y
Rueckert, D
Bai, W
Item Type: Conference Paper
Abstract: In recent years, convolutional neural networks have demonstrated promising performance in a variety of medical image segmentation tasks. However, when a trained segmentation model is deployed into the real clinical world, the model may not perform optimally. A major challenge is the potential poor-quality segmentations generated due to degraded image quality or domain shift issues. There is a timely need to develop an automated quality control method that can detect poor segmentations and feedback to clinicians. Here we propose a novel deep generative model-based framework for quality control of cardiac MRI segmentation. It first learns a manifold of good-quality image-segmentation pairs using a generative model. The quality of a given test segmentation is then assessed by evaluating the difference from its projection onto the good-quality manifold. In particular, the projection is refined through iterative search in the latent space. The proposed method achieves high prediction accuracy on two publicly available cardiac MRI datasets. Moreover, it shows better generalisation ability than traditional regression-based methods. Our approach provides a real-time and model-agnostic quality control for cardiac MRI segmentation, which has the potential to be integrated into clinical image analysis workflows.
Issue Date: 29-Sep-2020
Date of Acceptance: 1-Jun-2020
URI: http://hdl.handle.net/10044/1/80378
DOI: 10.1007/978-3-030-59719-1_9
ISSN: 0302-9743
Publisher: Springer Verlag
Start Page: 88
End Page: 97
Journal / Book Title: Lecture Notes in Computer Science
Volume: 12264
Copyright Statement: © 2020 Springer Nature Switzerland AG. The final publication is available at Springer via https://doi.org/10.1007/978-3-030-59719-1_9
Sponsor/Funder: Engineering & Physical Science Research Council (EPSRC)
Funder's Grant Number: EP/P001009/1
Conference Name: International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI)
Keywords: eess.IV
Artificial Intelligence & Image Processing
Notes: The paper is accepted to MICCAI 2020
Publication Status: Published
Start Date: 2020-10-04
Finish Date: 2020-10-08
Conference Place: Lima, Peru
Online Publication Date: 2020-09-29
Appears in Collections:Computing
Faculty of Medicine
Department of Brain Sciences
Faculty of Engineering