Machine Learning for Medical Image Reconstruction.pdf

发布时间：2022-05-31 发布人：admin 分类：说明书资料大小：36.00M 资料格式：pdf 举报版权申诉

zztq-10666065-16359647490793830833.pdf-第1页.png

第1页 / 共161页

zztq-10666065-16359647490793830833.pdf-第2页.png

第2页 / 共161页

zztq-10666065-16359647490793830833.pdf-第3页.png

第3页 / 共161页

zztq-10666065-16359647490793830833.pdf-第4页.png

第4页 / 共161页

zztq-10666065-16359647490793830833.pdf-第5页.png

第5页 / 共161页

zztq-10666065-16359647490793830833.pdf-第6页.png

第6页 / 共161页

zztq-10666065-16359647490793830833.pdf-第7页.png

第7页 / 共161页

zztq-10666065-16359647490793830833.pdf-第8页.png

第8页 / 共161页

Preface

Organization

Contents

Deep Learning for Magnetic Resonance Imaging

Deep Learning Super-Resolution Enables Rapid Simultaneous Morphological and Quantitative Magnetic Resonance Imaging

1 Introduction

1.1 Background

1.2 Motivation

2 Related Work

3 Methods

3.1 Imaging Methodology

3.2 Transfer Learning Training for Dual-Contrast DESS

4 Results

5 Discussion and Conclusion

References

ETER-net: End to End MR Image Reconstruction Using Recurrent Neural Network

1 Introduction

2 Method

2.1 Network Architectures

2.2 Training Environment

2.3 Quantitative Evaluation

3 Results

4 Discussion

References

Cardiac MR Motion Artefact Correction from K-space Using Deep Learning-Based Reconstruction

1 Introduction

2 Background

3 Methods

3.1 Network Architecture

3.2 Implementation Details

4 Experimental Results

4.1 K-space Corruption for Synthetic Data

4.2 Quantitative Results on Synthetic Dataset

4.3 Qualitative Results on Real Motion Artefact Case

5 Discussion and Conclusion

References

Complex Fully Convolutional Neural Networks for MR Image Reconstruction

1 Introduction

2 Methodology

2.1 Problem Formulation

2.2 Network Architecture

2.3 Model Learning and Optimization

3 Results and Discussion

3.1 Experimental Settings and Evaluation

3.2 Results

4 Conclusion and Future Work

References

Magnetic Resonance Fingerprinting Reconstruction via Spatiotemporal Convolutional Neural Networks

1 Introduction

2 Materials and Methods

2.1 MRF and Parametric Map Acquisition

2.2 Spatiotemporal CNN MRF Reconstruction

2.3 Evaluation

3 Results

4 Discussion and Conclusion

References

Improved Time-Resolved MRA Using k-Space Deep Learning

1 Introduction

2 Theory

2.1 Problem Formulation

2.2 From ALOHA to Deep Neural Network

3 Method

4 Result

5 Conclusion

References

Joint Motion Estimation and Segmentation from Undersampled Cardiac MR Image

1 Introduction

1.1 Related Work

2 Methods

2.1 Unsupervised Cardiac Motion Estimation from Undersampled MR Image

2.2 Joint Cardiac Motion Estimation and Segmentation from Undersampled MR Image

3 Experiments and Results

4 Conclusion

References

Bayesian Deep Learning for Accelerated MR Image Reconstruction

1 Introduction

2 Methods

3 Experiments and Results

4 Discussion and Conclusion

References

Deep Learning for Computed Tomography

Sparse-View CT Reconstruction Using Wasserstein GANs

1 Introduction

2 Method

2.1 Experimental Setup

3 Results

4 Discussion and Conclusion

References

Detecting Anatomical Landmarks for Motion Estimation in Weight-Bearing Imaging of Knees

1 Introduction

2 Method

2.1 X-Ray Invariant Anatomical Landmark Detection

2.2 Training

2.3 Landmark Estimation

3 Experiments and Results

4 Conclusion and Outlook

References

A U-Nets Cascade for Sparse View Computed Tomography

1 Introduction

1.1 Sparse View Computed Tomography

2 Proposed Network Architecture

2.1 Data Consistency Layer

2.2 U-Nets Cascade

3 Numerical Experiments

3.1 Dataset

3.2 Network Architectures and Training

3.3 Conclusion

References

Deep Learning for General Image Reconstruction

Approximate k-Space Models and Deep Learning for Fast Photoacoustic Reconstruction

1 Introduction

2 Forward and Inverse Models

2.1 Photoacoustic Tomography

2.2 Fast Approximate Forward and Inverse Models

3 Learned Reconstruction with Approximate Models

3.1 Learned Iterative Reconstruction

3.2 An Iterative Gradient Network

4 Computational Results for In-Vivo Measurements

4.1 Data Acquisition and Preparation

4.2 Training of Proposed Network

4.3 Reconstructions of In-Vivo Measurements

4.4 Discussion

5 Conclusions

References

Deep Learning Based Image Reconstruction for Diffuse Optical Tomography

1 Introduction

2 Methodology

2.1 Generating Training Data for DOT Reconstruction

2.2 Reconstructing Images from DOT Measurements

3 Experiments and Results

4 Conclusion

References

Image Reconstruction via Variational Network for Real-Time Hand-Held Sound-Speed Imaging

1 Introduction

2 Methods

2.1 Variational Network

3 Results

4 Discussion

References

Towards Arbitrary Noise Augmentation—Deep Learning for Sampling from Arbitrary Probability Distributions

1 Introduction

2 Conventional Sampling Methods

2.1 Inversion Sampling

2.2 Rejection Sampling

2.3 Mixture of Gaussians

2.4 Markov-Chain-Monte-Carlo

2.5 FCNN Sampling

3 Results

4 Conclusion

References

Left Atria Reconstruction from a Series of Sparse Catheter Paths Using Neural Networks

1 Introduction and Related Work

2 Methods

2.1 Reconstruction Scenarios: Using Sphere Intersection vs An Atria Path

3 Experiments and Results

3.1 Sphere Intersection

3.2 Synthetic Catheter Path Reconstruction

3.3 Laboratory Phantom

4 Conclusions and Future Work

References

High Quality Ultrasonic Multi-line Transmission Through Deep Learning

1 Introduction

2 Methods

3 Experimental Evaluation

4 Conclusion

References

Author Index

Florian Knoll Andreas Maier Daniel Rueckert (Eds.) Machine Learning for Medical Image Reconstruction 4 7 0 1 1 S C N L First International Workshop, MLMIR 2018 Held in Conjunction with MICCAI 2018 Granada, Spain, September 16, 2018 Proceedings 123

Lecture Notes in Computer Science 11074 Commenced Publication in 1973 Founding and Former Series Editors: Gerhard Goos, Juris Hartmanis, and Jan van Leeuwen Editorial Board David Hutchison Lancaster University, Lancaster, UK Takeo Kanade Carnegie Mellon University, Pittsburgh, PA, USA Josef Kittler University of Surrey, Guildford, UK Jon M. Kleinberg Cornell University, Ithaca, NY, USA Friedemann Mattern ETH Zurich, Zurich, Switzerland John C. Mitchell Stanford University, Stanford, CA, USA Moni Naor Weizmann Institute of Science, Rehovot, Israel C. Pandu Rangan Indian Institute of Technology Madras, Chennai, India Bernhard Steffen TU Dortmund University, Dortmund, Germany Demetri Terzopoulos University of California, Los Angeles, CA, USA Doug Tygar University of California, Berkeley, CA, USA Gerhard Weikum Max Planck Institute for Informatics, Saarbrücken, Germany

More information about this series at http://www.springer.com/series/7412

Florian Knoll Andreas Maier Daniel Rueckert (Eds.) Machine Learning for Medical Image Reconstruction First International Workshop, MLMIR 2018 Held in Conjunction with MICCAI 2018 Granada, Spain, September 16, 2018 Proceedings 123

Editors Florian Knoll New York University New York, NY USA Andreas Maier University of Erlangen-Nuremberg Erlangen Germany Daniel Rueckert Imperial College London London UK ISSN 0302-9743 Lecture Notes in Computer Science ISBN 978-3-030-00128-5 https://doi.org/10.1007/978-3-030-00129-2 ISBN 978-3-030-00129-2 (eBook) ISSN 1611-3349 (electronic) Library of Congress Control Number: 2018953025 LNCS Sublibrary: SL6 – Image Processing, Computer Vision, Pattern Recognition, and Graphics © Springer Nature Switzerland AG 2018 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, speciﬁcally the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microﬁlms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a speciﬁc statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional afﬁliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Preface We are proud to present the proceedings of the First Workshop on Machine Learning for Medical Image Reconstruction (MLMIR), which was held on 16th September 2018 in Granada, Spain, as part of the 21st Medical Image Computing and Computer Assisted Intervention (MICCAI) conference. in machine transform-based learning. Whereas Image reconstruction is currently undergoing a paradigm shift that is driven by or advances traditionally optimization-based methods have dominated methods for image reconstruction, machine learning has opened up the opportunity for new data-driven approaches, which have demonstrated a number of advantages over traditional approaches. In particular, deep learning techniques have shown signiﬁcant potential for image reconstruction and offer interesting new approaches. Finally, machine learning approaches also offer the possibility of application-speciﬁc image reconstruction, e.g., in motion-compensated cardiac or fetal imaging. This is supported by the success of machine learning in other inverse problems by multiple groups worldwide, with encouraging results and increasing interest. Coinci- dentally, this year is the centenary of the Radon transform and the 250th anniversary of the birth of Joseph Fourier. The Fourier transform and the Radon transform provide the mathematical foundation for tomography and medical imaging. It seems appro- priate and timely to consider how to invert the Radon transform and Fourier transform via machine learning, and have this workshop serve as a forum to reﬂect this emerging trend of image reconstruction research. In this respect, it will frame a fresh new way to recharge or redeﬁne the reconstruction algorithms with extensive prior knowledge for superior diagnostic performance. The aim of the workshop was to drive scientiﬁc discussion of advanced machine learning techniques for image acquisition and image reconstruction, opportunities for new applications, as well as challenges in the evaluation and validation of ML-based reconstruction approaches. We were fortunate that Jong Chul Ye (KAIST) and Michael Unser (EPFL) gave fascinating keynote lectures that summarised the state of the art in this emerging ﬁeld. Finally, we received 21 submissions and were able to accept 17 papers for inclusion in the workshop. The topics of the accepted papers cover the full range of medical image reconstruction problems, and deep learning dominates the machine learning approaches that are used to tackle the reconstruction problems. July 2018 Florian Knoll Andreas Maier Daniel Rueckert

Organization Workshop Organizers Daniel Rueckert Florian Knoll Andreas Maier Imperial College London, UK New York University, USA University of Erlangen, Germany Scientiﬁc Programme Committee Bernhard Kainz Bho Zhu Bruno De Man Claudia Prieto Dong Liang Enhao Gong Essam Rashed Ge Wang Greg Zaharchuk Guang Yang Jo Schlemper Jonas Adler Jong Chul Ye Jose Caballero Jo Hajnal Joseph Cheng Mariappan Nadar Matthew Rosen Michiel Schaap Morteza Mardani Ozan Öktem Rebecca Fahrig Simon Arridge Thomas Pock Tobias Wuerﬂ Imperial College London, UK Havard University, USA GE, USA King’s College London, UK Chinese Academy of Sciences, China Stanford University, USA British University in Egypt, Egypt Rensselaer Polytechnic Institute, USA Stanford University, USA Royal Brompton Hospital, UK Imperial College London, UK Royal Institute of Technology, Sweden KAIST, South Korea Twitter, UK King’s College London, UK Stanford University, USA Siemens Healthcare, USA Havard University, USA HeartFlow, USA Stanford University, USA Royal Institute of Technology, Sweden Siemens Healthcare, Germany University College London, UK Graz University of Technology, Austria Friedrich-Alexander-University Erlangen-Nuremberg, Tolga Cukur Bilkent University, Turkey Germany

Contents Deep Learning for Magnetic Resonance Imaging Deep Learning Super-Resolution Enables Rapid Simultaneous Morphological and Quantitative Magnetic Resonance Imaging . . . . . . . . . . . Akshay Chaudhari, Zhongnan Fang, Jin Hyung Lee, Garry Gold, and Brian Hargreaves ETER-net: End to End MR Image Reconstruction Using Recurrent Neural Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Changheun Oh, Dongchan Kim, Jun-Young Chung, Yeji Han, and HyunWook Park Cardiac MR Motion Artefact Correction from K-space Using Deep Learning-Based Reconstruction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ilkay Oksuz, James Clough, Aurelien Bustin, Gastao Cruz, Claudia Prieto, Rene Botnar, Daniel Rueckert, Julia A. Schnabel, and Andrew P. King Complex Fully Convolutional Neural Networks for MR Image Reconstruction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Muneer Ahmad Dedmari, Sailesh Conjeti, Santiago Estrada, Phillip Ehses, Tony Stöcker, and Martin Reuter Magnetic Resonance Fingerprinting Reconstruction via Spatiotemporal Convolutional Neural Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fabian Balsiger, Amaresha Shridhar Konar, Shivaprasad Chikop, Vimal Chandran, Olivier Scheidegger, Sairam Geethanath, and Mauricio Reyes 3 12 21 30 39 Improved Time-Resolved MRA Using k-Space Deep Learning . . . . . . . . . . . 47 Eunju Cha, Eung Yeop Kim, and Jong Chul Ye Joint Motion Estimation and Segmentation from Undersampled Cardiac MR Image . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 Chen Qin, Wenjia Bai, Jo Schlemper, Steffen E. Petersen, Stefan K. Piechnik, Stefan Neubauer, and Daniel Rueckert Bayesian Deep Learning for Accelerated MR Image Reconstruction . . . . . . . 64 Jo Schlemper, Daniel C. Castro, Wenjia Bai, Chen Qin, Ozan Oktay, Jinming Duan, Anthony N. Price, Jo Hajnal, and Daniel Rueckert

分享到：

赞收藏

资料库

Machine Learning for Medical Image Reconstruction.pdf

相关推荐

人工智能

热门标签

最新资料