Search

Convolutional Neural Networks (CNNs) have shown remarkable progress in medical image segmentation. However, lesion segmentation remains a challenge to state-of-the-art CNN-based algorithms due to the variance in scales and shapes. On the one hand, tiny lesions are hard to be delineated precisely from the medical images which are often of low resolutions. On the other hand, segmenting large-size lesions requires large receptive fields, which exacerbates the first challenge. In this paper, we present a scale-aware super-resolution network to adaptively segment lesions of various sizes from the low-resolution medical images. Our proposed network contains dual branches to simultaneously conduct lesion mask super-resolution and lesion image super-resolution. The image super-resolution branch will provide more detailed features for the segmentation branch, i.e., the mask super-resolution branch, for fine-grained segmentation. Meanwhile, we introduce scale-aware dilated convolution blocks into the multi-task decoders to adaptively adjust the receptive fields of the convolutional kernels according to the lesion sizes. To guide the segmentation branch to learn from richer high-resolution features, we propose a feature affinity module and a scale affinity module to enhance the multi-task learning of the dual branches. On multiple challenging lesion segmentation datasets, our proposed network achieved consistent improvements compared to other state-of-the-art methods., Comment: Journal paper under review. 10 pages. The first two authors contributed equally

Recently, physics-informed neural networks (PINNs) have emerged as a promising method for solving partial differential equations (PDEs). In this study, we establish a deep learning computational framework, HL-nets, for computing the flow field of hydrodynamic lubrication involving cavitation effects. Two classical cavitation conditions, i.e., the Swift-Stieber (SS) condition and the Jakobsson-Floberg-Olsson (JFO) condition, are implemented in the PINNs to solve the Reynolds equation. For the non-negativity constraint of the SS cavitation condition, a penalizing scheme with a residual of the non-negativity and an imposing scheme with a continuous differentiable non-negative function are proposed. For the complementarity constraint of the JFO cavitation condition, the pressure and cavitation fraction are taken as the neural network outputs, and the residual of the Fischer-Burmeister (FB) equation constrains their complementary relationships. Multi-task learning (MTL) methods are applied to balance the newly introduced loss terms described above. To estimate the accuracy of HL-nets, we present a numerical solution of the Reynolds equation for oil-lubricated bearings involving cavitation. The results indicate that the proposed HL-nets can highly accurately simulate hydrodynamic lubrication involving cavitation phenomena. The imposing scheme can effectively improve the accuracy of the training results of PINNs, and it is expected to have great potential to be applied to different fields where the non-negativity constraint is needed.

Ranking and selection (R&S) is a popular model for studying discrete-event dynamic systems. It aims to select the best design (the design with the largest mean performance) from a finite set, where the mean of each design is unknown and has to be learned by samples. Great research efforts have been devoted to this problem in the literature for developing procedures with superior empirical performance and showing their optimality. In these efforts, myopic procedures were popular. They select the best design using a 'naive' mechanism of iteratively and myopically improving an approximation of the objective measure. Although they are based on simple heuristics and lack theoretical support, they turned out highly effective, and often achieved competitive empirical performance compared to procedures that were proposed later and shown to be asymptotically optimal. In this paper, we theoretically analyze these myopic procedures and prove that they also satisfy the optimality conditions of R&S, just like some other popular R&S methods. It explains the good performance of myopic procedures in various numerical tests, and provides good insight into the structure and theoretical development of efficient R&S procedures.

Ordinal optimization (OO) is a widely-studied technique for optimizing discrete-event dynamic systems (DEDS). It evaluates the performance of the system designs in a finite set by sampling and aims to correctly make ordinal comparison of the designs. A well-known method in OO is the optimal computing budget allocation (OCBA). It builds the optimality conditions for the number of samples allocated to each design, and the sample allocation that satisfies the optimality conditions is shown to asymptotically maximize the probability of correct selection for the best design. In this paper, we investigate two popular OCBA algorithms. With known variances for samples of each design, we characterize their convergence rates with respect to different performance measures. We first demonstrate that the two OCBA algorithms achieve the optimal convergence rate under measures of probability of correct selection and expected opportunity cost. It fills the void of convergence analysis for OCBA algorithms. Next, we extend our analysis to the measure of cumulative regret, a main measure studied in the field of machine learning. We show that with minor modification, the two OCBA algorithms can reach the optimal convergence rate under cumulative regret. It indicates the potential of broader use of algorithms designed based on the OCBA optimality conditions.

With the rapid development of aerospace technology, the vibration problem of the spacecraft flexible structure urgently needs to be solved. Magnetic fluids are a type of multi-functional smart materials, which can be employed in shock absorbers to eliminate these vibrations. Referring to the calculation methods of stiffness coefficients of other passive dampers, the stiffness coefficient formula of magnetic fluid shock absorbers (MFSAs) was derived. Meanwhile, a novel stiffness adjustable magnetic fluid shock absorber (SA-MFSA) was proposed. On the basis of the second-order buoyancy principle, a series of SA-MFSAs were fabricated. The range of stiffness coefficients covered by these SA-MFSAs contains the optimal stiffness coefficient estimated by formulas. The repulsive force measurement and vibration attenuation experiments were conducted on these SA-MFSAs. In the case of small amplitude, the relationship between the repulsive force and the offset distance was linear. The simulation and experiment curves of repulsive forces were in good agreement. The results of vibration attenuation experiments demonstrated that the rod length and the magnetic fluid mass influence the damping efficiency of SA-MFSAs. In addition, these results verified that the SA-MFSA with the optimal stiffness coefficient performed best. Therefore, the stiffness coefficient formula can guide the design of MFSAs., Comment: 18 pages, 12 figures

The knowledge gradient (KG) algorithm is a popular and effective algorithm for the best arm identification (BAI) problem. Due to the complex calculation of KG, theoretical analysis of this algorithm is difficult, and existing results are mostly about the asymptotic performance of it, e.g., consistency, asymptotic sample allocation, etc. In this research, we present new theoretical results about the finite-time performance of the KG algorithm. Under independent and normally distributed rewards, we derive bounds for the sample allocation of the algorithm. With these bounds, existing asymptotic results become simple corollaries. Furthermore, we derive upper and lower bounds for the probability of error and simple regret of the algorithm, and show the performance of the algorithm for the multi-armed bandit (MAB) problem. These developments not only extend the existing analysis of the KG algorithm, but can also be used to analyze other improvement-based algorithms. Last, we use numerical experiments to compare the bounds we derive and the performance of the KG algorithm.

Computing the grain boundary (GB) counterparts to bulk phase diagrams represents an emerging research direction. Using a classical embrittlement model system Ga-doped Al alloy, this study demonstrates the feasibility of computing temperature- and composition-dependent GB diagrams to represent not only equilibrium thermodynamic and structural characters, but also mechanical properties. Specifically, hybrid Monte Carlo and molecular dynamics (MC/MD) simulations are used to obtain the equilibrium GB structure as a function of temperature and composition. Simulated GB structures are validated by aberration-corrected scanning transmission electron microscopy. Subsequently, MD tensile tests are performed on the simulated equilibrium GB structures. GB diagrams are computed for not only GB adsorption and structural disorder, but also interfacial structural and chemical widths, MD ultimate tensile strength, and MD tensile toughness. This study suggests a research direction to investigate GB composition-structure-property relationships via computing GB diagrams of thermodynamic, structural, and mechanical (or potentially other) properties.

The novel coronavirus disease 2019 (COVID-19) characterized by atypical pneumonia has caused millions of deaths worldwide. Automatically segmenting lesions from chest Computed Tomography (CT) is a promising way to assist doctors in COVID-19 screening, treatment planning, and follow-up monitoring. However, voxel-wise annotations are extremely expert-demanding and scarce, especially when it comes to novel diseases, while an abundance of unlabeled data could be available. To tackle the challenge of limited annotations, in this paper, we propose an uncertainty-guided dual-consistency learning network (UDC-Net) for semi-supervised COVID-19 lesion segmentation from CT images. Specifically, we present a dual-consistency learning scheme that simultaneously imposes image transformation equivalence and feature perturbation invariance to effectively harness the knowledge from unlabeled data. We then quantify the segmentation uncertainty in two forms and employ them together to guide the consistency regularization for more reliable unsupervised learning. Extensive experiments showed that our proposed UDC-Net improves the fully supervised method by 6.3% in Dice and outperforms other competitive semi-supervised approaches by significant margins, demonstrating high potential in real-world clinical practice., Comment: Accepted to MICCAI2021. The first two authors contributed equally. Code is available at https://github.com/poiuohke/UDC-Net