Acoustic Forward Model for Guided Wave Propagation and Scattering in a Pipe Bend

The sections of pipe bends are hot spots for wall thinning due to accelerated corrosion by fluid flow. Conventionally, the thickness of a bend wall is evaluated by local point-by-point ultrasonic measurement, which is slow and costly. Guided wave tomography is an attractive method that enables the monitoring of a whole bend area by processing the waves excited and received by transducer arrays. The main challenge associated with the tomography of the bend is the development of an appropriate forward model, which should simply and efficiently handle the wave propagation in a complex bend model. In this study, we developed a two-dimensional (2D) acoustic forward model to replace the complex three-dimensional (3D) bend domain with a rectangular domain that is made artificially anisotropic by using Thomsen parameters. Thomsen parameters allow the consideration of the directional dependence of the velocity of the wave in the model. Good agreement was found between predictions and experiments performed on a 220 mm diameter (d) pipe with 1.5d bend radius, including the wave-field focusing effect and the steering effect of scattered wave-fields from defects.

AIEADA 1.0: Efficient high-dimensional variational data assimilation with machine-learned reduced-order models

Data assimilation (DA) in the geophysical sciences remains the cornerstone of robust forecasts from numerical models. Indeed, DA plays a crucial role in the quality of numerical weather prediction, and is a crucial building block that has allowed dramatic improvements in weather forecasting over the past few decades. DA is commonly framed in a variational setting, where one solves an optimization problem within a Bayesian formulation using raw model forecasts as a prior, and observations as likelihood. This leads to a DA objective function that needs to be minimized, where the decision variables are the initial conditions specified to the model. In traditional DA, the forward model is numerically and computationally expensive. Here we replace the forward model with a low-dimensional, data-driven, and differentiable emulator. Consequently, gradients of our DA objective function with respect to the decision variables are obtained rapidly via automatic differentiation. We demonstrate our approach by performing an emulator-assisted DA forecast of geopotential height. Our results indicate that emulator-assisted DA is faster than traditional equation-based DA forecasts by four orders of magnitude, allowing computations to be performed on a workstation rather than a dedicated high-performance computer. In addition, we describe accuracy benefits of emulator-assisted DA when compared to simply using the emulator for forecasting (i.e., without DA). Our overall formulation is denoted AIAEDA (Artificial Intelligence Emulator Assisted Data Assimilation).

Neural Underpinning of Japanese Particle Processing in Non-native Speakers

Grammar acquisition by non-native learners (L2) is typically less successful and may produce fundamentally different grammatical systems than that by native speakers (L1). The neural representation of grammatical processing between L1 and L2 speakers remains controversial. We hypothesized that working memory is the primary source of L1/L2 differences, and operationalized working memory is an active inference within the predictive coding account, which models grammatical processes as higher-level neuronal representations of cortical hierarchies, generating predictions (forward model) of lower-level representations. A functional MRI study was conducted with L1 Japanese speakers and highly proficient Japanese learners requiring oral production of grammatically correct Japanese particles. Selecting proper particles requires forward model-dependent active inference as their functions are highly context-dependent. As a control, participants read out a visually designated mora indicated by underlining. Particle selection by L1/L2 groups commonly activated the bilateral inferior frontal gyrus/insula, pre-supplementary motor area, left caudate, middle temporal gyrus, and right cerebellum, which constituted the core linguistic production system. In contrast, the left inferior frontal sulcus, known as the neural substrate of verbal working memory, showed more prominent activation in L2 than in L1. Thus, the active inference process causes L1/L2 differences even in highly proficient L2 learners.

Neural underpinning of Japanese particle processing in non-native speakers

Grammar acquisition by non-native learners (L2) is typically less successful and may produce fundamentally different grammatical systems than that by native speakers (L1). The neural representation of grammatical processing between L1 and L2 speakers remains controversial. We hypothesized that working memory is the primary source of L1/L2 differences, and operationalized working memory is an active inference within the predictive coding account, which models grammatical processes as higher-level neuronal representations of cortical hierarchies, generating predictions (forward model) of lower-level representations. A functional MRI study was conducted with L1 Japanese speakers and highly proficient Japanese learners requiring oral production of grammatically correct Japanese particles. Selecting proper particles requires forward model-dependent active inference as their functions are highly context-dependent. As a control, participants read out a visually designated mora indicated by underlining. Particle selection by L1/L2 groups commonly activated the bilateral inferior frontal gyrus/insula, pre-supplementary motor area, left caudate, middle temporal gyrus, and right cerebellum, which constituted the core linguistic production system. In contrast, the left inferior frontal sulcus, known as the neural substrate of verbal working memory, showed more prominent activation in L2 than in L1. Thus, the active inference process causes L1/L2 differences even in highly proficient L2 learners.

A Non-Iterative Method Combined with Neural Network Embedded in Physical Model to Solve the Imaging of Electromagnetic Inverse Scattering Problem

The main purpose of this paper is to solve the electromagnetic inverse scattering problem (ISP). Compared with conventional tomography technology, it considers the interaction between the internal structure of the scene and the electromagnetic wave in a more realistic manner. However, due to the nonlinearity of ISP, the conventional calculation scheme usually has some problems, such as the unsatisfactory imaging effect and high computational cost. To solve these problems and improve the imaging quality, this paper presents a simple method named the diagonal matrix inversion method (DMI) to estimate the distribution of scatterer contrast (DSC) and a Generative Adversarial Network (GAN) which could optimize the DSC obtained by DMI and make it closer to the real distribution of scatterer contrast. In order to make the distribution of scatterer contrast generated by GAN more accurate, the forward model is embedded in the GAN. Moreover, because of the existence of the forward model, not only is the DSC generated by the generator similar to the original distribution of the scatterer contrast in the numerical distribution, but the numerical of each point is also approximate to the original.

A reference tissue forward model for improved PET accuracy using within-scan displacement studies

We report a novel forward-model implementation of the full reference tissue model (fFTRM) that addresses the fast-exchange approximation employed by the simplified reference tissue model (SRTM) by incorporating a non-zero dissociation time constant from the specifically bound compartment. The forward computational approach avoided errors associated with noisy and nonorthogonal basis functions using an inverse linear model. Compared to analysis by a multilinear single-compartment reference tissue model (MRTM), fFTRM provided improved accuracy for estimation of binding potentials at early times in the scan, with no worse reproducibility across sessions. To test the model’s ability to identify small focal changes in binding potential using a within-scan challenge, we employed a nonhuman primate model of focal dopamine release elicited by deep brain microstimulation remote to ventral striatum (VST) during imaging by simultaneous PET and fMRI. The new model reported an unambiguously lateralized response in VST consistent with fMRI, whereas the MRTM-derived response was not lateralized and was consistent with simulations of model bias. The proposed model enabled better accuracy in PET [11C]raclopride displacement studies and may also facilitate challenges sooner after injection, thereby recovering some sensitivity lost to radioactive decay of the PET tracer.

Supplemental Material: Zircon (U-Th)/He thermochronology of Grand Canyon resolves 1250 Ma unroofing at the Great Unconformity and <20 Ma canyon carving

All geologic and thermochronologic constraints used in the forward models, major forward model results from the zircon radiation damage accumulation and annealing model (ZRDAAM) of Guenthner et al. (2013), and alternative HeFTy forward model result using different sample inputs.<br>

Physics Augmented U-Net: A High-Frequency Aware Generative Prior for Microscopy

A key challenge in optical microscopy is to image fast at high-resolution. To address this problem, we propose "Physics Augmented U-Net", which combines deep learning and structured illumination microscopy (SIM). In SIM, the structured illumination aliases out-of-band high-frequencies to the passband of the microscope; thus SIM captures some high-frequencies even when the image is sampled at low-resolution. To utilize these features, we propose a three-element method: 1) a modified U-Net model, 2) a physics-based forward model of SIM 3) an inference algorithm combining the two models. The modified U-Net architecture is similar to the seminal work, but the bottleneck is modified by concatenating two latent vectors, one encoding low-frequencies (LFLV), and the other encoding high-frequencies (HFLV). LFLV is learned by U-Net contracting path, and HFLV is learned by a second encoding path. In the inference mode, the high-frequency encoder is removed; HFLV is then optimized to fit the measured microscopy images to the output of the forward model for the generated image by the U-Net. We validated our method on two different datasets under different experimental conditions. Since a latent vector is optimized instead of a 2D image, the inference mode is less computationally complex. The proposed model is also more stable compared to other generative prior-based methods. Finally, as the forward model is independent of the U-Net, Physics Augmented U-Net can enhance resolution on any variation of SIM without further retraining.

Reconstruction of Depolarization Pattern in Myocardial Infarction Cases Using Two Cascaded Stages of Artificial Neural Networks

Our paper introduces a new technique for diagnosis of various heart diseases without the need of highly experts to investigate the electrocardiogram (ECG). Using the same electrodes of the ECG machine, it will be able to transmit directly the electrical activity inside the heart to a moving picture. Our technique is based on artificial intelligence algorithm using artificial neural networks (ANN). Finding the trans-membrane potential (TMP) inside the heart from the body surface potential (BSP) is known as the inverse problem of ECG. To have a unique solution for the inverse problem the data used should be obtained from a forward model. A three dimensional (3-D) model of cellular activation whole heart embedded in torso is simulated and solved using COMSOL Multiphysics software. In our previous paper, one ANN succeeded in displaying the wave propagation on the surface of a normal heart. In this paper, we used a configuration of ANNs to display different cases of heart with myocardial infarction (MI). To check the system accuracy, eight MI cases with different sizes and locations in the heart are simulated in the forward model. This configuration proved to be highly accurate in displaying each MI case -size and location- presenting the infarction as an area with no electrical activity.