In this paper, we propose a novel deep convolutional neural network (CNN)-based algorithm for solving ill-posed inverse problems. Regularized iterative algorithms have emerged as the standard approach to ill-posed inverse problems in the past few decades. These methods produce excellent results, but can be challenging to deploy in practice due to factors including the high computational cost of the forward and adjoint operators and the difficulty of hyperparameter selection. The starting point of this paper is the observation that unrolled iterative methods have the form of a CNN (filtering followed by pointwise non-linearity) when the normal operator ( $H^{*}H$ , where $H^{*}$ is the adjoint of the forward imaging operator, $H$ ) of the forward model is a convolution. Based on this observation, we propose using direct inversion followed by a CNN to solve normal-convolutional inverse problems. The direct inversion encapsulates the physical model of the system, but leads to artifacts when the problem is ill posed; the CNN combines multiresolution decomposition and residual learning in order to learn to remove these artifacts while preserving image structure. We demonstrate the performance of the proposed network in sparse-view reconstruction (down to 50 views) on parallel beam X-ray computed tomography in synthetic phantoms as well as in real experimental sinograms. The proposed network outperforms total variation-regularized iterative reconstruction for the more realistic phantoms and requires less than a second to reconstruct a $512\times 512$ image on the GPU.

/pdf/deep-convolutional-neural-network-for-inverse-problems-in-20ejy5u0ob.pdf

Deep Convolutional Neural Network for Inverse Problems in Imaging

Compressed sensing (CS) utilizes the sparsity of magnetic resonance (MR) images to enable accurate reconstruction from undersampled k-space data. Recent CS methods have employed analytical sparsifying transforms such as wavelets, curvelets, and finite differences. In this paper, we propose a novel framework for adaptively learning the sparsifying transform (dictionary), and reconstructing the image simultaneously from highly undersampled k-space data. The sparsity in this framework is enforced on overlapping image patches emphasizing local structure. Moreover, the dictionary is adapted to the particular image instance thereby favoring better sparsities and consequently much higher undersampling rates. The proposed alternating reconstruction algorithm learns the sparsifying dictionary, and uses it to remove aliasing and noise in one step, and subsequently restores and fills-in the k-space data in the other step. Numerical experiments are conducted on MR images and on real MR data of several anatomies with a variety of sampling schemes. The results demonstrate dramatic improvements on the order of 4-18 dB in reconstruction error and doubling of the acceptable undersampling factor using the proposed adaptive dictionary as compared to previous CS methods. These improvements persist over a wide range of practical data signal-to-noise ratios, without any parameter tuning.

MR Image Reconstruction From Highly Undersampled k-Space Data by Dictionary Learning

Summary Background Deep learning offers considerable promise for medical diagnostics. We aimed to evaluate the diagnostic accuracy of deep learning algorithms versus health-care professionals in classifying diseases using medical imaging. Methods In this systematic review and meta-analysis, we searched Ovid-MEDLINE, Embase, Science Citation Index, and Conference Proceedings Citation Index for studies published from Jan 1, 2012, to June 6, 2019. Studies comparing the diagnostic performance of deep learning models and health-care professionals based on medical imaging, for any disease, were included. We excluded studies that used medical waveform data graphics material or investigated the accuracy of image segmentation rather than disease classification. We extracted binary diagnostic accuracy data and constructed contingency tables to derive the outcomes of interest: sensitivity and specificity. Studies undertaking an out-of-sample external validation were included in a meta-analysis, using a unified hierarchical model. This study is registered with PROSPERO, CRD42018091176. Findings Our search identified 31 587 studies, of which 82 (describing 147 patient cohorts) were included. 69 studies provided enough data to construct contingency tables, enabling calculation of test accuracy, with sensitivity ranging from 9·7% to 100·0% (mean 79·1%, SD 0·2) and specificity ranging from 38·9% to 100·0% (mean 88·3%, SD 0·1). An out-of-sample external validation was done in 25 studies, of which 14 made the comparison between deep learning models and health-care professionals in the same sample. Comparison of the performance between health-care professionals in these 14 studies, when restricting the analysis to the contingency table for each study reporting the highest accuracy, found a pooled sensitivity of 87·0% (95% CI 83·0–90·2) for deep learning models and 86·4% (79·9–91·0) for health-care professionals, and a pooled specificity of 92·5% (95% CI 85·1–96·4) for deep learning models and 90·5% (80·6–95·7) for health-care professionals. Interpretation Our review found the diagnostic performance of deep learning models to be equivalent to that of health-care professionals. However, a major finding of the review is that few studies presented externally validated results or compared the performance of deep learning models and health-care professionals using the same sample. Additionally, poor reporting is prevalent in deep learning studies, which limits reliable interpretation of the reported diagnostic accuracy. New reporting standards that address specific challenges of deep learning could improve future studies, enabling greater confidence in the results of future evaluations of this promising technology. Funding None.

A comparison of deep learning performance against health-care professionals in detecting diseases from medical imaging: a systematic review and meta-analysis

Thank you for downloading magnetic resonance imaging physical principles and sequence design. As you may know, people have look numerous times for their chosen books like this magnetic resonance imaging physical principles and sequence design, but end up in harmful downloads. Rather than enjoying a good book with a cup of coffee in the afternoon, instead they juggled with some malicious bugs inside their laptop.

Magnetic Resonance Imaging Physical Principles And Sequence Design

What has happened in machine learning lately, and what does it mean for the future of medical image analysis? Machine learning has witnessed a tremendous amount of attention over the last few years. The current boom started around 2009 when so-called deep artificial neural networks began outperforming other established models on a number of important benchmarks. Deep neural networks are now the state-of-the-art machine learning models across a variety of areas, from image analysis to natural language processing, and widely deployed in academia and industry. These developments have a huge potential for medical imaging technology, medical data analysis, medical diagnostics and healthcare in general, slowly being realized. We provide a short overview of recent advances and some associated challenges in machine learning applied to medical image processing and image analysis. As this has become a very broad and fast expanding field we will not survey the entire landscape of applications, but put particular focus on deep learning in MRI. Our aim is threefold: (i) give a brief introduction to deep learning with pointers to core references; (ii) indicate how deep learning has been applied to the entire MRI processing chain, from acquisition to image retrieval, from segmentation to disease prediction; (iii) provide a starting point for people interested in experimenting and perhaps contributing to the field of deep learning for medical imaging by pointing out good educational resources, state-of-the-art open-source code, and interesting sources of data and problems related medical imaging.

An overview of deep learning in medical imaging focusing on MRI

Introduction Balanced steady state free precession (bSSFP) imaging has an established role in the CMR assessment of ventricular function and wall motion [1]. However, measured bSSFP signals do not always agree with simple theoretical predictions. More accurate signal models would enable optimization of imaging parameters (e.g. prescribed flip angle and TR) in order to maximize signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR). In this work, we examine the effects of (A) non-ideal slice profile, (B) off-resonance, and (C) radiofrequency transmit (B1+) variation, to identify a model that accurately predicts myocardial signal behavior in bSSFP, over the range of imaging parameters routinely used at 3 T.

/pdf/50-myocardial-signal-behaviors-of-balanced-ssfp-imaging-at-3-4qfs33odi8.pdf

50 Myocardial signal behaviors of balanced SSFP imaging at 3 T

Clinical magnetic resonance imaging (MRI) of the pituitary gland sometimes fails to detect small pituitary tumors due to limited signal-to-noise ratio (SNR) and spatial resolution. Thus, neurosurgeons may need to resort to surgical exploration and systematic slicing of the pituitary gland to identify the small pituitary tumors. In this work, we designed a single-loop miniature flexible coil that can be surgically positioned millimeters from the pituitary gland, enabling high-SNR pituitary MRI. We investigated the spatial distributions of the image SNR of the miniature coil, via both numerical simulation and phantom experiments. We also explored the feasibility of increased SNR within the pituitary gland based on simulated surgical placements. Compared to the commercial head coil, our miniature coil achieved up to a 19-fold SNR improvement within the region of interest, and the simulation and phantom experiment reached a good agreement, with an error of 1.1% ± 0.8%. High resolution MRI scans further demonstrated the visual improvement of the miniature coil against the commercial head coil. The cross-validation of the simulation and the phantom experiment showed the potential of using the numerical simulation model to accelerate the coil design prototyping and iteration and to optimize coil design in the future.

A Miniature Flexible Coil for High-SNR MRI of the Pituitary Gland

/pdf/1091-b1-non-uniformity-in-3-t-cmr-patient-study-14rwq29k2s.pdf

1091 B1+ non-uniformity in 3 T CMR: patient study

Abstract Context Gestational diabetes (GDM) imposes long-term adverse health effects on the mother and fetus. The role of magnetic resonance imaging (MRI) during early gestation in GDM has not been well-studied. Objective To investigate the role of quantitative MRI measurements of placental volume and perfusion, with distribution of maternal adiposity, during early gestation in GDM. Methods At UCLA outpatient antenatal obstetrics clinics, ∼200 pregnant women recruited in the first trimester were followed temporally through pregnancy until parturition. Two placental MRI scans were prospectively performed at 14 to 16 weeks and 19 to 24 weeks gestational age (GA). Placental volume and blood flow (PBF) were calculated from placental regions of interest; maternal adiposity distribution was assessed by subcutaneous fat area ratio (SFAR) and visceral fat area ratio (VFAR). Statistical comparisons were performed using the two-tailed t test. Predictive logistic regression modeling was evaluated by area under the curve (AUC). Results Of a total 186 subjects, 21 subjects (11.3%) developed GDM. VFAR was higher in GDM vs the control group, at both time points (P < 0.001 each). Placental volume was greater in GDM vs the control group at 19 to 24 weeks GA (P = 0.01). Combining VFAR, placental volume and perfusion, improved the AUC to 0.83 at 14 to 16 weeks (positive predictive value [PPV] = 0.77, negative predictive value [NPV] = 0.83), and 0.81 at 19 to 24 weeks GA (PPV = 0.73, NPV = 0.86). Conclusion A combination of MRI-based placental volume, perfusion, and visceral adiposity during early pregnancy demonstrates significant changes in GDM and provides a proof of concept for predicting the subsequent development of GDM.

/pdf/utility-of-in-vivo-magnetic-resonance-imaging-is-predictive-1wcdivx5.pdf

Kyunghyun Sung

Papers

50 Myocardial signal behaviors of balanced SSFP imaging at 3 T

A Miniature Flexible Coil for High-SNR MRI of the Pituitary Gland

1091 B1+ non-uniformity in 3 T CMR: patient study

Utility of In Vivo Magnetic Resonance Imaging Is Predictive of Gestational Diabetes Mellitus During Early Pregnancy