Thomas Küstner1,2,3, Karim Armanious1,2, Jiahuan Yang1, Bin Yang1, Fritz Schick2, Sergios Gatidis2. 1. Department for Signal Processing and System Theory, University of Stuttgart, Stuttgart, Germany. 2. Diagnostic and Interventional Radiology, University Hospital Tuebingen, Tuebingen, Germany. 3. School of Biomedical Engineering & Imaging Sciences, King's College London, St. Thomas' Hospital, London, United Kingdom.
Abstract
PURPOSE: Motion is 1 extrinsic source for imaging artifacts in MRI that can strongly deteriorate image quality and, thus, impair diagnostic accuracy. In addition to involuntary physiological motion such as respiration and cardiac motion, intended and accidental patient movements can occur. Any impairment by motion artifacts can reduce the reliability and precision of the diagnosis and a motion-free reacquisition can become time- and cost-intensive. Numerous motion correction strategies have been proposed to reduce or prevent motion artifacts. These methods have in common that they need to be applied during the actual measurement procedure with a-priori knowledge about the expected motion type and appearance. For retrospective motion correction and without the existence of any a-priori knowledge, this problem is still challenging. METHODS: We propose the use of deep learning frameworks to perform retrospective motion correction in a reference-free setting by learning from pairs of motion-free and motion-affected images. For this image-to-image translation problem, we propose and compare a variational auto encoder and generative adversarial network. Feasibility and influences of motion type and optimal architecture are investigated by blinded subjective image quality assessment and by quantitative image similarity metrics. RESULTS: We observed that generative adversarial network-based motion correction is feasible producing near-realistic motion-free images as confirmed by blinded subjective image quality assessment. Generative adversarial network-based motion correction accordingly resulted in images with high evaluation metrics (normalized root mean squared error <0.08, structural similarity index >0.8, normalized mutual information >0.9). CONCLUSION: Deep learning-based retrospective restoration of motion artifacts is feasible resulting in near-realistic motion-free images. However, the image translation task can alter or hide anatomical features and, therefore, the clinical applicability of this technique has to be evaluated in future studies.
PURPOSE: Motion is 1 extrinsic source for imaging artifacts in MRI that can strongly deteriorate image quality and, thus, impair diagnostic accuracy. In addition to involuntary physiological motion such as respiration and cardiac motion, intended and accidental patient movements can occur. Any impairment by motion artifacts can reduce the reliability and precision of the diagnosis and a motion-free reacquisition can become time- and cost-intensive. Numerous motion correction strategies have been proposed to reduce or prevent motion artifacts. These methods have in common that they need to be applied during the actual measurement procedure with a-priori knowledge about the expected motion type and appearance. For retrospective motion correction and without the existence of any a-priori knowledge, this problem is still challenging. METHODS: We propose the use of deep learning frameworks to perform retrospective motion correction in a reference-free setting by learning from pairs of motion-free and motion-affected images. For this image-to-image translation problem, we propose and compare a variational auto encoder and generative adversarial network. Feasibility and influences of motion type and optimal architecture are investigated by blinded subjective image quality assessment and by quantitative image similarity metrics. RESULTS: We observed that generative adversarial network-based motion correction is feasible producing near-realistic motion-free images as confirmed by blinded subjective image quality assessment. Generative adversarial network-based motion correction accordingly resulted in images with high evaluation metrics (normalized root mean squared error <0.08, structural similarity index >0.8, normalized mutual information >0.9). CONCLUSION: Deep learning-based retrospective restoration of motion artifacts is feasible resulting in near-realistic motion-free images. However, the image translation task can alter or hide anatomical features and, therefore, the clinical applicability of this technique has to be evaluated in future studies.
Authors: H Huang; J H Siewerdsen; W Zbijewski; C R Weiss; M Unberath; T Ehtiati; A Sisniega Journal: Phys Med Biol Date: 2022-06-16 Impact factor: 4.174
Authors: Ben A Duffy; Lu Zhao; Farshid Sepehrband; Joyce Min; Danny Jj Wang; Yonggang Shi; Arthur W Toga; Hosung Kim Journal: Neuroimage Date: 2021-01-15 Impact factor: 6.556
Authors: Mohamad Abdi; Xue Feng; Changyu Sun; Kenneth C Bilchick; Craig H Meyer; Frederick H Epstein Journal: Magn Reson Med Date: 2021-05-22 Impact factor: 3.737
Authors: Akshay S Chaudhari; Christopher M Sandino; Elizabeth K Cole; David B Larson; Garry E Gold; Shreyas S Vasanawala; Matthew P Lungren; Brian A Hargreaves; Curtis P Langlotz Journal: J Magn Reson Imaging Date: 2020-08-24 Impact factor: 5.119
Authors: Anouk S Verschuur; Vivian Boswinkel; Chantal M W Tax; Jochen A C van Osch; Ingrid M Nijholt; Cornelis H Slump; Linda S de Vries; Gerda van Wezel-Meijler; Alexander Leemans; Martijn F Boomsma Journal: J Neuroimaging Date: 2022-03-07 Impact factor: 2.324
Authors: Jennifer A Steeden; Michael Quail; Alexander Gotschy; Kristian H Mortensen; Andreas Hauptmann; Simon Arridge; Rodney Jones; Vivek Muthurangu Journal: J Cardiovasc Magn Reson Date: 2020-08-03 Impact factor: 5.364