☆ 4.7 Article

Physics-Driven Synthetic Data Learning for Biomedical Magnetic Resonance: The imaging physics-based data synthesis paradigm for artificial intelligence

IEEE SIGNAL PROCESSING MAGAZINE (2023)

相关参考文献

注意：仅列出部分参考文献，下载原文获取全部文献信息。

Article Computer Science, Artificial Intelligence

A Sparse Model-Inspired Deep Thresholding Network for Exponential Signal Reconstruction--Application in Fast Biological Spectroscopy

Zi Wang et al.

Summary: In this work, we propose a DL architecture called MoDern for spectra reconstruction from undersampled data by combining the advantages of the sparse model-based optimization method and data-driven DL. Extensive experiments on synthetic and biological data demonstrate that MoDern achieves more robust, high-fidelity, and ultrafast reconstruction compared to existing methods. Remarkably, MoDern has a small number of network parameters and generalizes well to biological data in various scenarios. Furthermore, we extend MoDern to a user-friendly cloud computing platform (XCloud-MoDern), contributing a promising strategy for the further development of biological applications.

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS (2023)

添加到收藏夹

Article Computer Science, Artificial Intelligence

DeepSPIO: Super Paramagnetic Iron Oxide Particle Quantification Using Deep Learning in Magnetic Resonance Imaging

Gabriel Della Maggiora et al.

Summary: In this study, a new Deep Learning method was proposed to quantify the distribution of SPIO concentration. The method utilized a novel data acquisition sequence and multiple decoders to improve gradient flow. The accuracy of the method was demonstrated through tests with simulated images and actual scan data.

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE (2022)

添加到收藏夹

Article Radiology, Nuclear Medicine & Medical Imaging

Ultrafast water-fat separation using deep learning-based single-shot MRI

Xinran Chen et al.

Summary: The study proposed a deep learning-based reconstruction method for simultaneous water and fat image acquisition with spatiotemporally encoded MRI, demonstrating better fidelity and higher spatial resolution compared to existing methods.

MAGNETIC RESONANCE IN MEDICINE (2022)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Physics-informed neural networks for myocardial perfusion MRI quantification

Rudolf L.M. van Herten et al.

MEDICAL IMAGE ANALYSIS (2022)

添加到收藏夹

Article Computer Science, Interdisciplinary Applications

Model-Based SyntheTic Data-Driven Learning (MOST-DL): Application in Single-Shot T2 Mapping With Severe Head Motion Using Overlapping-Echo Acquisition

Qinqin Yang et al.

Summary: The study proposes a framework, MOST-DL, to address the lack of training data in supervised learning scenarios for motion-robust T-2 mapping. By using synthetic data and realistic textures, the proposed method significantly reduces ghosting and motion artifacts in T-2 maps, showing potential for application in motion-prone patients.

IEEE TRANSACTIONS ON MEDICAL IMAGING (2022)

添加到收藏夹

Article Computer Science, Interdisciplinary Applications

Physics-Informed Neural Networks for Brain Hemodynamic Predictions Using Medical Imaging

Mohammad Sarabian et al.

Summary: This study proposes a physics-informed deep learning framework that utilizes a reduced-order model simulation to generate high-resolution brain hemodynamic parameters, augmenting sparse clinical measurements. The framework is validated against in vivo velocity measurements obtained from MRI scans and shows potential in diagnosing cerebral vasospasm.

IEEE TRANSACTIONS ON MEDICAL IMAGING (2022)

添加到收藏夹

Article Radiology, Nuclear Medicine & Medical Imaging

Artificial neural network for multi-echo gradient echo-based myelin water fraction estimation

Soozy Jung et al.

Summary: The study demonstrated robust myelin water fraction (MWF) mapping using an artificial neural network (ANN) with multi-echo gradient-echo (GRE) signal. Both numerical simulations and in vivo data showed high accuracy and reduced noise with the ANN compared to the nonlinear least-squares method. These results suggest that the ANN facilitates more robust MWF mapping in multi-echo gradient-echo sequences.

MAGNETIC RESONANCE IN MEDICINE (2021)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Using synthetic data generation to train a cardiac motion tag tracking neural network

Michael Loecher et al.

Summary: A CNN-based method for cardiac MRI tag tracking was developed and validated, showing accurate and precise results when tested on both analytical phantom and in vivo data. This study demonstrates the effectiveness of automated tag tracking using synthetic data trained CNN model.

MEDICAL IMAGE ANALYSIS (2021)

添加到收藏夹

Article Radiology, Nuclear Medicine & Medical Imaging

MRzero - Automated discovery of MRI sequences using supervised learning

A. Loktyushin et al.

Summary: A supervised learning framework is proposed for automatically generating MR sequences and reconstruction based on target contrast, allowing for efficient exploration of novel strategies. The scanning and reconstruction process is simulated using RF and gradient moment events, and delay times, with optimization based on data fidelity, SAR penalty, and scan time. The approach demonstrates the potential for automated MR sequence generation through differentiable Bloch equation simulations and supervised learning.

MAGNETIC RESONANCE IN MEDICINE (2021)

添加到收藏夹

Editorial Material Engineering, Biomedical

Synthetic data in machine learning for medicine and healthcare

Richard J. Chen et al.

Summary: The proliferation of synthetic data in artificial intelligence for medicine and healthcare has raised concerns about the vulnerabilities of the software and the challenges faced by current policies.

NATURE BIOMEDICAL ENGINEERING (2021)

添加到收藏夹

Article Computer Science, Interdisciplinary Applications

Separation of Metabolites and Macromolecules for Short-TE 1H-MRSI Using Learned Component-Specific Representations

Yahang Li et al.

Summary: A new approach integrating imaging physics and representation learning is introduced to solve the separation problem of metabolite and macromolecule signals in short-TE data. By using deep autoencoders to learn low-dimensional representations and a constrained reconstruction formulation, an efficient algorithm is developed to effectively separate the signals. Simulation and experimental results demonstrate the capability of the proposed method in practical H-1-MRSI data.

IEEE TRANSACTIONS ON MEDICAL IMAGING (2021)

添加到收藏夹

Article Chemistry, Multidisciplinary

Accelerated Nuclear Magnetic Resonance Spectroscopy with Deep Learning

Xiaobo Qu et al.

ANGEWANDTE CHEMIE-INTERNATIONAL EDITION (2020)

添加到收藏夹

Editorial Material Engineering, Electrical & Electronic

Computational MRI: Compressive Sensing and Beyond

Mathews Jacob et al.

IEEE SIGNAL PROCESSING MAGAZINE (2020)

添加到收藏夹

Article Computer Science, Interdisciplinary Applications

Constrained Magnetic Resonance Spectroscopic Imaging by Learning Nonlinear Low-Dimensional Models

Fan Lam et al.

IEEE TRANSACTIONS ON MEDICAL IMAGING (2020)

添加到收藏夹

Article Radiology, Nuclear Medicine & Medical Imaging

A Transfer-Learning Approach for Accelerated MRI Using Deep Neural Networks

Salman Ul Hassan Dar et al.

MAGNETIC RESONANCE IN MEDICINE (2020)

添加到收藏夹

Article Multidisciplinary Sciences

In vivo imaging of phosphocreatine with artificial neural networks

Lin Chen et al.

NATURE COMMUNICATIONS (2020)

添加到收藏夹

Article Radiology, Nuclear Medicine & Medical Imaging

DeepcomplexMRI: Exploiting deep residual network for fast parallel MR imaging with complex convolution

Shanshan Wang et al.

MAGNETIC RESONANCE IMAGING (2020)

添加到收藏夹

Article Radiology, Nuclear Medicine & Medical Imaging

Deep learning-based reconstruction of in vivo pelvis conductivity with a 3D patch-based convolutional neural network trained on simulated MR data

Soraya Gavazzi et al.

MAGNETIC RESONANCE IN MEDICINE (2020)

添加到收藏夹

Article Neurosciences

Exploring linearity of deep neural network trained QSM: QSMnet

Woojin Jung et al.

NEUROIMAGE (2020)

添加到收藏夹

Article Radiology, Nuclear Medicine & Medical Imaging

Deblurring for spiral real-time MRI using convolutional neural networks

Yongwan Lim et al.

MAGNETIC RESONANCE IN MEDICINE (2020)

添加到收藏夹

Article Radiology, Nuclear Medicine & Medical Imaging

Intact metabolite spectrum mining by deep learning in proton magnetic resonance spectroscopy of the brain

Hyeong Hun Lee et al.

MAGNETIC RESONANCE IN MEDICINE (2019)

添加到收藏夹

Article Computer Science, Interdisciplinary Applications

Robust Single-Shot T2 Mapping via Multiple Overlapping-Echo Acquisition and Deep Neural Network

Jun Zhang et al.

IEEE TRANSACTIONS ON MEDICAL IMAGING (2019)

添加到收藏夹

Article Computer Science, Interdisciplinary Applications

Simulation and Synthesis in Medical Imaging

Alejandro F. Frangi et al.

IEEE TRANSACTIONS ON MEDICAL IMAGING (2018)

添加到收藏夹

Article Computer Science, Interdisciplinary Applications

Unsupervised Reverse Domain Adaptation for Synthetic Medical Images via Adversarial Training

Faisal Mahmood et al.

IEEE TRANSACTIONS ON MEDICAL IMAGING (2018)

添加到收藏夹

Article Radiology, Nuclear Medicine & Medical Imaging

MR fingerprinting Deep RecOnstruction NEtwork (DRONE)

Ouri Cohen et al.

MAGNETIC RESONANCE IN MEDICINE (2018)

添加到收藏夹

Article Radiology, Nuclear Medicine & Medical Imaging

Single-shot T2 mapping using overlapping-echo detachment planar imaging and a deep convolutional neural network

Congbo Cai et al.

MAGNETIC RESONANCE IN MEDICINE (2018)

添加到收藏夹

Article Computer Science, Interdisciplinary Applications

Fast Realistic MRI Simulations Based on Generalized Multi-Pool Exchange Tissue Model

Fang Liu et al.

IEEE TRANSACTIONS ON MEDICAL IMAGING (2017)

添加到收藏夹

© Peeref 2019-2024. All rights reserved.