Lack of ground-truth MR images impedes the common supervised training of neural networks for image reconstruction. To cope with this challenge, this article leverages unpaired adversarial training for reconstruction networks, where the inputs are undersampled k-space and naively reconstructed images from one dataset, and the labels are high-quality images from another dataset. The reconstruction networks consist of a generator which suppresses the input image artifacts, and a discriminator using a pool of (unpaired) labels to adjust the reconstruction quality. The generator is an unrolled neural network - a cascade of convolutional and data consistency layers. The discriminator is also a multilayer CNN that plays the role of a critic scoring the quality of reconstructed images based on the Wasserstein distance. Our experiments with knee MRI datasets demonstrate that the proposed unpaired training enables diagnostic-quality reconstruction when high-quality image labels are not available for the input types of interest, or when the amount of labels is small. In addition, our adversarial training scheme can achieve better image quality (as rated by expert radiologists) compared with the paired training schemes with pixel-wise loss.

The location information of entities in construction sites, such as workers and construction machines, is valuable in project management and safety. Therefore, as nonintrusive and accurate solutions, various vision-based methods have been proposed to track entities in construction sites and obtain their three-dimensional (3D) coordinates. However, most existing vision-based methods realize 3D localizations by basing entity matching on the epipolar line, which brings instability in entity matching due to the calculation error of the epipolar line or failure to match entities when multiple entities are located on the same epipolar. To solve this problem, a novel framework based on reidentification is proposed to automatically match workers across two camera views, thereby obtaining their 3D coordinates in construction sites. In this framework, deep-learning-based computer vision algorithms are firstly used to detect and track workers in two camera views. Then, the reidentification (ReID) algorithm is applied to utilize tracked workers’ visual features to match the workers across both two camera views and different frames. As a result, for every matched pair, the worker’s pixel locations in two camera views can be obtained to calculate the 3D coordinates through triangulation. The implementation of videos recorded from a real construction project proves the feasibility and accuracy of this framework. Specifically, through utilizing the ReID algorithm to match workers, the framework achieves competitive results on workers matching with precision, recall, and accuracy of more than 99%, 93%, and 93%. Furthermore, it also effectively addresses the practical problems of ID repetition and ID switching. Meanwhile, this paper extends the application scenarios of reidentification algorithms in construction sites, thereby contributing to the future application of multiple-camera vision-based methods in construction sites.

Manual prescription of the field of view (FOV) by MRI technologists is variable and prolongs the scanning process. Often, the FOV is too large or crops critical anatomy. We propose a deep learning framework, trained by radiologists’ supervision, for automating FOV prescription. An intra-stack shared feature extraction network and an attention network are used to process a stack of 2D image inputs to generate scalars defining the location of a rectangular region of interest (ROI). The attention mechanism is used to make the model focus on a small number of informative slices in a stack. Then, the smallest FOV that makes the neural network predicted ROI free of aliasing is calculated by an algebraic operation derived from MR sampling theory. The framework’s performance is examined quantitatively with intersection over union (IoU) and pixel error on position and qualitatively with a reader study. The proposed model achieves an average IoU of 0.867 and an average ROI position error of 9.06 out of 512 pixels on 80 test cases, significantly better than two baseline models and not significantly different from a radiologist. Finally, the FOV given by the proposed framework achieves an acceptance rate of 92% from an experienced radiologist.

In recent years, extensive research has shown that deep learning-based compressed image reconstruction algorithms can achieve faster and better high-quality reconstruction for single-pixel imaging, and that reconstruction quality can be further improved by joint optimization of sampling and reconstruction. However, these network-based models mostly adopt end-to-end learning, and their structures are not interpretable. In this paper, we propose SRMU-Net, a sampling and reconstruction jointly optimized model unfolding network. A fully connected layer or a large convolutional layer that simulates compressed reconstruction is added to the compressed reconstruction network, which is composed of multiple cascaded iterative shrinkage thresholding algorithm (ISTA) unfolding iteration blocks. To achieve joint optimization of sampling and reconstruction, a specially designed network structure is proposed so that the sampling matrix can be input into ISTA unfolding iteration blocks as a learnable parameter. We have shown that the proposed network outperforms the existing algorithms by extensive simulations and experiments.