Automatic Inter-Frame Patient Motion Correction for Dynamic Cardiac PET Using Deep Learning

Patient motion during dynamic PET imaging can induce errors in myocardial blood flow (MBF) estimation. Motion correction for dynamic cardiac PET is challenging because the rapid tracer kinetics of 82Rb leads to substantial tracer distribution change across different dynamic frames over time, which can cause difficulties for image registration-based motion correction, particularly for early dynamic frames. In this paper, we developed an automatic deep learning-based motion correction (DeepMC) method for dynamic cardiac PET. In this study we focused on the detection and correction of inter-frame rigid translational motion caused by voluntary body movement and pattern change of respiratory motion. A bidirectional-3D LSTM network was developed to fully utilize both local and nonlocal temporal information in the 4D dynamic image data for motion detection. The network was trained and evaluated over motion-free patient scans with simulated motion so that the motion ground-truths are available, where one million samples based on 65 patient scans were used in training, and 600 samples based on 20 patient scans were used in evaluation. The proposed method was also evaluated using additional 10 patient datasets with real motion. We demonstrated that the proposed DeepMC obtained superior performance compared to conventional registration-based methods and other convolutional neural networks (CNN), in terms of motion estimation and MBF quantification accuracy. Once trained, DeepMC is much faster than the registration-based methods and can be easily integrated into the clinical workflow. In the future work, additional investigation is needed to evaluate this approach in a clinical context with realistic patient motion.

[1]  Hui Liu,et al.  Deep learning-based attenuation map generation for myocardial perfusion SPECT , 2020, European Journal of Nuclear Medicine and Molecular Imaging.

[2]  Luyao Shi,et al.  Automatic Diagnosis of Pulmonary Embolism Using an Attention-guided Framework: A Large-scale Study , 2020, MIDL.

[3]  Richard E Carson,et al.  Direct List Mode Parametric Reconstruction for Dynamic Cardiac SPECT , 2020, IEEE Transactions on Medical Imaging.

[4]  G. Fakhri,et al.  MR-based cardiac and respiratory motion correction of PET: application to static and dynamic cardiac 18F-FDG imaging , 2019, Physics in medicine and biology.

[5]  Chi Liu,et al.  Deep Learning based Respiratory Pattern Classification and Applications in PET/CT Motion Correction , 2019, 2019 IEEE Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC).

[6]  Luyao Shi,et al.  A Novel Loss Function Incorporating Imaging Acquisition Physics for PET Attenuation Map Generation using Deep Learning , 2019, MICCAI.

[7]  Tianyu Ma,et al.  An investigation of quantitative accuracy for deep learning based denoising in oncological PET , 2019, Physics in medicine and biology.

[8]  Huazhong Shu,et al.  Domain Progressive 3D Residual Convolution Network to Improve Low-Dose CT Imaging , 2019, IEEE Transactions on Medical Imaging.

[9]  Tim Mulnix,et al.  Data-driven voluntary body motion detection and non-rigid event-by-event correction for static and dynamic PET , 2019, Physics in medicine and biology.

[10]  Pingkun Yan,et al.  Deep learning in medical image registration: a survey , 2019, Machine Vision and Applications.

[11]  Adam M Alessio,et al.  Patient body motion correction for dynamic cardiac PET‐CT by attenuation‐emission alignment according to projection consistency conditions , 2019, Medical physics.

[12]  Nicolas A Karakatsanis,et al.  Correction of respiratory and cardiac motion in cardiac PET/MR using MR-based motion modeling , 2018, Physics in medicine and biology.

[13]  Yifan Wu,et al.  Repetitive Motion Estimation Network: Recover cardiac and respiratory signal from thoracic imaging , 2018, ArXiv.

[14]  Benjamin C. Lee,et al.  Automated dynamic motion correction using normalized gradient fields for 82rubidium PET myocardial blood flow quantification , 2018, Journal of Nuclear Cardiology.

[15]  Guang Li,et al.  CT Super-Resolution GAN Constrained by the Identical, Residual, and Cycle Learning Ensemble (GAN-CIRCLE) , 2018, IEEE Transactions on Medical Imaging.

[16]  Ge Wang,et al.  Structurally-Sensitive Multi-Scale Deep Neural Network for Low-Dose CT Denoising , 2018, IEEE Access.

[17]  Yun Dong,et al.  Self-Gating: An Adaptive Center-of-Mass Approach for Respiratory Gating in PET , 2018, IEEE Transactions on Medical Imaging.

[18]  K. Takeda,et al.  Usefulness of abdominal belt for restricting respiratory cardiac motion and improving image quality in myocardial perfusion PET , 2018, Journal of Nuclear Cardiology.

[19]  Venkatesh L. Murthy,et al.  Blood pool and tissue phase patient motion effects on 82rubidium PET myocardial blood flow quantification , 2018, Journal of Nuclear Cardiology.

[20]  Deniz Erdogmus,et al.  Real-Time Deep Pose Estimation With Geodesic Loss for Image-to-Template Rigid Registration , 2018, IEEE Transactions on Medical Imaging.

[21]  Dinggang Shen,et al.  Medical Image Synthesis with Deep Convolutional Adversarial Networks , 2018, IEEE Transactions on Biomedical Engineering.

[22]  Tim Mulnix,et al.  Respiratory Motion Compensation for PET/CT with Motion Information Derived from Matched Attenuation-Corrected Gated PET Data , 2018, The Journal of Nuclear Medicine.

[23]  C. Meinel,et al.  Deep Learning for Medical Image Analysis , 2017, ArXiv.

[24]  Heng Tao Shen,et al.  Video Captioning With Attention-Based LSTM and Semantic Consistency , 2017, IEEE Transactions on Multimedia.

[25]  Xiao Jin,et al.  Data-driven event-by-event respiratory motion correction using TOF PET list-mode centroid of distribution , 2017, Physics in medicine and biology.

[26]  Piotr J. Slomka,et al.  Enhancing Cardiac PET by Motion Correction Techniques , 2017, Current Cardiology Reports.

[27]  A. Sinusas,et al.  Quantification of myocardial blood flow with 82Rb: Validation with 15O-water using time-of-flight and point-spread-function modeling , 2016, EJNMMI Research.

[28]  R. deKemp,et al.  Patient motion effects on the quantification of regional myocardial blood flow with dynamic PET imaging. , 2016, Medical physics.

[29]  Dit-Yan Yeung,et al.  Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting , 2015, NIPS.

[30]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[31]  Navdeep Jaitly,et al.  Hybrid speech recognition with Deep Bidirectional LSTM , 2013, 2013 IEEE Workshop on Automatic Speech Recognition and Understanding.

[32]  Ran Klein,et al.  Does quantification of myocardial flow reserve using rubidium-82 positron emission tomography facilitate detection of multivessel coronary artery disease? , 2012, Journal of Nuclear Cardiology.

[33]  Francis R. Verdun,et al.  Quantification of myocardial blood flow with 82Rb positron emission tomography: clinical validation with 15O-water , 2012, European Journal of Nuclear Medicine and Molecular Imaging.

[34]  Jun Hatazawa,et al.  Effects of patient movement on measurements of myocardial blood flow and viability in resting 15O-water PET studies , 2012, Journal of Nuclear Cardiology.

[35]  D. Dey,et al.  Automatic 3D registration of dynamic stress and rest (82)Rb and flurpiridaz F 18 myocardial perfusion PET data for patient motion detection and correction. , 2011, Medical physics.

[36]  Paul E Kinahan,et al.  Attenuation-emission alignment in cardiac PET/CT based on consistency conditions. , 2010, Medical physics.

[37]  S. Hochreiter,et al.  Long Short-Term Memory , 1997, Neural Computation.

[38]  J. Paul Siebert,et al.  Learning Rigid Image Registration - Utilizing Convolutional Neural Networks for Medical Image Registration , 2018, BIOIMAGING.

[39]  Hollis G. Potter,et al.  Author Manuscript , 2013 .