Automated detection of motion artifacts in brain MR images using deep learning.

Autor: Manso Jimeno M; Department of Biomedical Engineering, Columbia University in the City of New York, New York, New York, USA.; Columbia Magnetic Resonance Research Center, Columbia University in the City of New York, New York, New York, USA., Ravi KS; Department of Biomedical Engineering, Columbia University in the City of New York, New York, New York, USA.; Columbia Magnetic Resonance Research Center, Columbia University in the City of New York, New York, New York, USA., Fung M; MR Clinical Solutions, GE Healthcare, New York, New York, USA., Oyekunle D; Department of Radiology, University College Hospital, Ibadan, Nigeria., Ogbole G; Department of Radiology, University College Hospital, Ibadan, Nigeria., Vaughan JT Jr; Department of Biomedical Engineering, Columbia University in the City of New York, New York, New York, USA.; Columbia Magnetic Resonance Research Center, Columbia University in the City of New York, New York, New York, USA.; Department of Radiology, Columbia University Medical Center, New York, New York, USA.; Zuckerman Institute, Columbia University in the City of New York, New York, New York, USA., Geethanath S; Columbia Magnetic Resonance Research Center, Columbia University in the City of New York, New York, New York, USA.; Department of Radiology and Radiological Science, John Hopkins University, Baltimore, Maryland, USA.
Jazyk: angličtina
Zdroj: NMR in biomedicine [NMR Biomed] 2025 Jan; Vol. 38 (1), pp. e5276. Date of Electronic Publication: 2024 Oct 22.
DOI: 10.1002/nbm.5276
Abstrakt: Quality assessment, including inspecting the images for artifacts, is a critical step during magnetic resonance imaging (MRI) data acquisition to ensure data quality and downstream analysis or interpretation success. This study demonstrates a deep learning (DL) model to detect rigid motion in T 1 -weighted brain images. We leveraged a 2D convolutional neural network (CNN) trained on motion-synthesized data for three-class classification and tested it on publicly available retrospective and prospective datasets. Grad-CAM heatmaps enabled the identification of failure modes and provided an interpretation of the model's results. The model achieved average precision and recall metrics of 85% and 80% on six motion-simulated retrospective datasets. Additionally, the model's classifications on the prospective dataset showed 93% agreement with the labeling of a radiologist a strong inverse correlation (-0.84) compared to average edge strength, an image quality metric indicative of motion. This model is aimed at inline automatic detection of motion artifacts, accelerating part of the time-consuming quality assessment (QA) process and augmenting expertise on-site, particularly relevant in low-resource settings where local MR knowledge is scarce.
(© 2024 John Wiley & Sons Ltd.)
Databáze: MEDLINE
Externí odkaz: