Calibration-Less Multi-Coil Compressed Sensing Magnetic Resonance Image Reconstruction Based on OSCAR Regularization

Over the last decade, the combination of compressed sensing (CS) with acquisition over multiple receiver coils in magnetic resonance imaging (MRI) has allowed the emergence of faster scans while maintaining a good signal-to-noise ratio (SNR). Self-calibrating techniques, such as ESPiRIT, have become the standard approach to estimating the coil sensitivity maps prior to the reconstruction stage. In this work, we proceed differently and introduce a new calibration-less multi-coil CS reconstruction method. Calibration-less techniques no longer require the prior extraction of sensitivity maps to perform multi-coil image reconstruction but usually alternate estimation sensitivity map estimation and image reconstruction. Here, to get rid of the nonconvexity of the latter approach we reconstruct as many MR images as the number of coils. To compensate for the ill-posedness of this inverse problem, we leverage structured sparsity of the multi-coil images in a wavelet transform domain while adapting to variations in SNR across coils owing to the OSCAR (octagonal shrinkage and clustering algorithm for regression) regularization. Coil-specific complex-valued MR images are thus obtained by minimizing a convex but nonsmooth objective function using the proximal primal-dual Condat-Vù algorithm. Comparison and validation on retrospective Cartesian and non-Cartesian studies based on the Brain fastMRI data set demonstrate that the proposed reconstruction method outperforms the state-of-the-art (ℓ1-ESPIRiT, calibration-less AC-LORAKS and CaLM methods) significantly on magnitude images for the T1 and FLAIR contrasts. Additionally, further validation operated on 8 to 20-fold prospectively accelerated high-resolution ex vivo human brain MRI data collected at 7 Tesla confirms the retrospective results. Overall, OSCAR-based regularization preserves phase information more accurately (both visually and quantitatively) compared to other approaches, an asset that can only be assessed on real prospective experiments.

Covid-19 detection via deep neural network and occlusion sensitivity maps

Deep learning approaches have attracted a lot of attention in the automatic detection of Covid-19 and transfer learning is the most common approach. However, majority of the pre-trained models are trained on color images, which can cause inefficiencies when fine-tuning the models on Covid-19 images which are often grayscale. To address this issue, we propose a deep learning architecture called CovidNet which requires a relatively smaller number of parameters. CovidNet accepts grayscale images as inputs and is suitable for training with limited training dataset. Experimental results show that CovidNet outperforms other state-of-the-art deep learning models for Covid-19 detection.

Covid-19 detection via deep neural network and occlusion sensitivity maps

Deep learning approaches have attracted a lot of attention in the automatic detection of Covid-19 and transfer learning is the most common approach. However, majority of the pre-trained models are trained on color images, which can cause inefficiencies when fine-tuning the models on Covid-19 images which are often grayscale. To address this issue, we propose a deep learning architecture called CovidNet which requires a relatively smaller number of parameters. CovidNet accepts grayscale images as inputs and is suitable for training with limited training dataset. Experimental results show that CovidNet outperforms other state-of-the-art deep learning models for Covid-19 detection.

WHAT IDENTIFIES A WHALE BY ITS FLUKE? ON THE BENEFIT OF INTERPRETABLE MACHINE LEARNING FOR WHALE IDENTIFICATION

Interpretable and explainable machine learning have proven to be promising approaches to verify the quality of a data-driven model in general as well as to obtain more information about the quality of certain observations in practise. In this paper, we use these approaches for an application in the marine sciences to support the monitoring of whales. Whale population monitoring is an important element of whale conservation, where the identification of whales plays an important role in this process, for example to trace the migration of whales over time and space. Classical approaches use photographs and a manual mapping with special focus on the shape of the whale flukes and their unique pigmentation. However, this is not feasible for comprehensive monitoring. Machine learning methods, especially deep neural networks, have shown that they can efficiently solve the automatic observation of a large number of whales. Despite their success for many different tasks such as identification, further potentials such as interpretability and their benefits have not yet been exploited. Our main contribution is an analysis of interpretation tools, especially occlusion sensitivity maps, and the question of how the gained insights can help a whale researcher. For our analysis, we use images of humpback whale flukes provided by the Kaggle Challenge ”Humpback Whale Identification”. By means of spectral cluster analysis of heatmaps, which indicate which parts of the image are important for a decision, we can show that the they can be grouped in a meaningful way. Moreover, it appears that characteristics automatically determined by a neural network correspond to those that are considered important by a whale expert.

Developing a Universal Mirror–mirror Laser Mapping System for Single Event Effect Research

Research on single event effects (SEEs) is significant to the design and manufacture of modern electronic devices. By applying two photon absorption (TPA) ultra-fast pulsed lasers, extra electron-hole pairs (EHPs) are generated in a desired location on a chip, simulating the process that could occur in the circuit by energetic particles. In this study, a SEE sensitivity mapping system is described which uses this method to generate real-time sensitivity maps for various electronic devices. The system hardware includes an attenuator to control the energy, a Pockels cell as a fast-optical switcher and a mirror–mirror module to project the laser beam into a certain location. The system software developed for this application controls the laser system, automatically generates sensitivity maps, communicates with the testing devices and logs the SEE results. The two main features of this laser mapping system are: high scanning velocity for large area scanning (about 1 × 1 mm) and high spatial resolution for small area scanning (about 1 × 1 μm). To verify this mapping system, sensitivity maps were generated for static random access memory (SRAM) built with 65 nm technology and for commercial operational amplifiers (op-amps). The achieved sensitivity maps were compared with circuitry analysis and laser testing results, confirming this mapping system to be effective.

Sensitivity maps for time-reverse imaging: an accuracy study for the Los Humeros Geothermal Field (Mexico)

SUMMARY The estimation of the source–location accuracy of microseismic events in reservoirs is of significant importance. Time-reverse imaging (TRI) provides a highly accurate localization scheme to locate events by time-reversing the recorded full wavefield and back propagating it through a velocity model. So far, the influence of the station geometry and the velocity model on the source–location accuracy is not well known. Therefore, sensitivity maps are developed using the geothermal site of Los Humeros in Mexico to evaluate the spatial variability of the source–location accuracy. Sensitivity maps are created with an assumed gradient velocity model with a constant vp–vs ratio and with a realistic velocity model for the region of Los Humeros. The positions of 27 stations deployed in Los Humeros from September 2017 to September 2018 are used as surface receivers. An automatic localization scheme is proposed that does not rely on any a priori information about the sources and thus negates any user bias in the source locations. The sensitivity maps are created by simulating numerous uniformly distributed sources simultaneously and locating these sources using TRI. The found source locations are compared to the initial source locations to estimate the achieved accuracy. The resulting sensitivity maps show that the station geometry introduces complex patterns in the spatial variation of accuracy. Furthermore, the influence of the station geometry on the source–location accuracy is larger than the influence of the velocity model. Finally, a microearthquake recorded at the geothermal site of Los Humeros is located to demonstrate the usability of the derived sensitivity maps. This study stresses the importance of optimizing station networks to enhance the accuracy when locating seismic events using TRI.