Joseph Y. Cheng

Undersampled magnetic resonance image (MRI) reconstruction is typically an ill-posed linear inverse task. The time and resource intensive computations require tradeoffs between accuracy and speed. In addition, state-of-the-art compressed sensing (CS) analytics are not cognizant of the image diagnostic quality. To address these challenges, we propose a novel CS framework that uses generative adversarial networks (GAN) to model the (low-dimensional) manifold of high-quality MR images. Leveraging a mixture of least-squares (LS) GANs and pixel-wise ℓ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> /ℓ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> cost, a deep residual network with skip connections is trained as the generator that learns to remove the aliasing artifacts by projecting onto the image manifold. The LSGAN learns the texture details, while the ℓ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> /ℓ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> cost suppresses high-frequency noise. A discriminator network, which is a multilayer convolutional neural network (CNN), plays the role of a perceptual cost that is then jointly trained based on high-quality MR images to score the quality of retrieved images. In the operational phase, an initial aliased estimate (e.g., simply obtained by zero-filling) is propagated into the trained generator to output the desired reconstruction. This demands a very low computational overhead. Extensive evaluations are performed on a large contrast-enhanced MR dataset of pediatric patients. Images rated by expert radiologists corroborate that GANCS retrieves higher quality images with improved fine texture details compared with conventional Wavelet-based and dictionary-learning-based CS schemes as well as with deep-learning-based schemes using pixel-wise training. In addition, it offers reconstruction times of under a few milliseconds, which are two orders of magnitude faster than the current state-of-the-art CS-MRI schemes.

Compressed sensing (CS) reconstruction methods leverage sparse structure in underlying signals to recover high-resolution images from highly undersampled measurements. When applied to magnetic resonance imaging (MRI), CS has the potential to dramatically shorten MRI scan times, increase diagnostic value, and improve overall patient experience. However, CS has several shortcomings which limit its clinical translation such as: 1) artifacts arising from inaccurate sparse modelling assumptions, 2) extensive parameter tuning required for each clinical application, and 3) clinically infeasible reconstruction times. Recently, CS has been extended to incorporate deep neural networks as a way of learning complex image priors from historical exam data. Commonly referred to as unrolled neural networks, these techniques have proven to be a compelling and practical approach to address the challenges of sparse CS. In this tutorial, we will review the classical compressed sensing formulation and outline steps needed to transform this formulation into a deep learning-based reconstruction framework. Supplementary open source code in Python will be used to demonstrate this approach with open databases. Further, we will discuss considerations in applying unrolled neural networks in the clinical setting.

In this study, a deep learning algorithm based on the you-only-look-once (YOLO) approach is proposed for the quality inspection of printed circuit boards (PCBs). The high accuracy and efficiency of deep learning algorithms has resulted in their increased adoption in every field. Similarly, accurate detection of defects in PCBs by using deep learning algorithms, such as convolutional neural networks (CNNs), has garnered considerable attention. In the proposed method, highly skilled quality inspection engineers first use an interface to record and label defective PCBs. The data are then used to train a YOLO/CNN model to detect defects in PCBs. In this study, 11,000 images and a network of 24 convolutional layers and 2 fully connected layers were used. The proposed model achieved a defect detection accuracy of 98.79% in PCBs with a batch size of 32.

PURPOSE: To develop and assess motion correction techniques for high-resolution pediatric abdominal volumetric magnetic resonance images acquired free-breathing with high scan efficiency. MATERIALS AND METHODS: First, variable-density sampling and radial-like phase-encode ordering were incorporated into the 3D Cartesian acquisition. Second, intrinsic multichannel butterfly navigators were used to measure respiratory motion. Lastly, these estimates are applied for both motion-weighted data-consistency in a compressed sensing and parallel imaging reconstruction, and for nonrigid motion correction using a localized autofocusing framework. With Institutional Review Board approval and informed consent/assent, studies were performed on 22 consecutive pediatric patients. Two radiologists independently scored the images for overall image quality, degree of motion artifacts, and sharpness of hepatic vessels and the diaphragm. The results were assessed using paired Wilcoxon test and weighted kappa coefficient for interobserver agreements. RESULTS: The complete procedure yielded significantly better overall image quality (mean score of 4.7 out of 5) when compared to using no correction (mean score of 3.4, P < 0.05) and to using motion-weighted accelerated imaging (mean score of 3.9, P < 0.05). With an average scan time of 28 seconds, the proposed method resulted in comparable image quality to conventional prospective respiratory-triggered acquisitions with an average scan time of 91 seconds (mean score of 4.5). CONCLUSION: With the proposed methods, diagnosable high-resolution abdominal volumetric scans can be obtained from free-breathing data acquisitions.

PURPOSE: Deep learning has had success with MRI reconstruction, but previously published works use real-valued networks. The few works which have tried complex-valued networks have not fully assessed their impact on phase. Therefore, the purpose of this work is to fully investigate end-to-end complex-valued convolutional neural networks (CNNs) for accelerated MRI reconstruction and in several phase-based applications in comparison to 2-channel real-valued networks. METHODS: Several complex-valued activation functions for MRI reconstruction were implemented, and their performance was compared. Complex-valued convolution was implemented and tested on an unrolled network architecture and a U-Net-based architecture over a wide range of network widths and depths with knee, body, and phase-contrast datasets. RESULTS: Quantitative and qualitative results demonstrated that complex-valued CNNs with complex-valued convolutions provided superior reconstructions compared to real-valued convolutions with the same number of trainable parameters for both an unrolled network architecture and a U-Net-based architecture, and for 3 different datasets. Complex-valued CNNs consistently had superior normalized RMS error, structural similarity index, and peak SNR compared to real-valued CNNs. CONCLUSION: Complex-valued CNNs can enable superior accelerated MRI reconstruction and phase-based applications such as fat-water separation, and flow quantification compared to real-valued convolutional neural networks.