Covid-19 refereeing duration and impact in major medical journals

Two partly conflicting academic pressures from the seriousness of the Covid-19 pandemic are the need for faster peer review of Covid-19 health-related research and greater scrutiny of its findings. This paper investigates whether decreases in peer review durations for Covid-19 articles were universal across 97 major medical journals, Nature, Science, and Cell. The results suggest that on average, Covid-19 articles submitted during 2020 were reviewed 1.7–2.1 times faster than non-Covid-19 articles submitted during 2017–2020. Nevertheless, whilst the review speed of Covid-19 research was particularly fast during the first five months (1.9–3.4 times faster) of the pandemic (January–May 2020), this speed advantage was no longer evident for articles submitted November–December 2020. Faster peer review also associates with higher citation impact for Covid-19 articles in the same journals, suggesting it did not usually compromise the scholarly impact of important Covid-19 research. Overall, then, it seems that core medical and general journals responded quickly but carefully to the pandemic, although the situation returned closer to normal within a year. Peer Review https://publons.com/publon/10.1162/qss_a_00176

Metal Surface Defect Detection Using Modified YOLO

Aiming at the problems of inefficient detection caused by traditional manual inspection and unclear features in metal surface defect detection, an improved metal surface defect detection technology based on the You Only Look Once (YOLO) model is presented. The shallow features of the 11th layer in the Darknet-53 are combined with the deep features of the neural network to generate a new scale feature layer using the basis of the network structure of YOLOv3. Its goal is to extract more features of small defects. Furthermore, then, K-Means++ is used to reduce the sensitivity to the initial cluster center when analyzing the size information of the anchor box. The optimal anchor box is selected to make the positioning more accurate. The performance of the modified metal surface defect detection technology is compared with other detection methods on the Tianchi dataset. The results show that the average detection accuracy of the modified YOLO model is 75.1%, which ia higher than that of YOLOv3. Furthermore, it also has a great detection speed advantage, compared with faster region-based convolutional neural network (Faster R-CNN) and other detection algorithms. The improved YOLO model can make the highly accurate location information of the small defect target and has strong real-time performance.

Spectral-Coding-Based Compressive Single-Pixel NIR Spectroscopy in the Sub-Millisecond Regime

In this contribution, we present a high-speed, multiplex, grating spectrometer based on a spectral coding approach that is founded on principles of compressive sensing. The spectrometer employs a single-pixel InGaAs detector to measure the signals encoded by an amplitude spatial light modulator (digital micromirror device, DMD). This approach leads to a speed advantage and multiplex sensitivity advantage atypical for standard dispersive systems. Exploiting the 18.2 kHz pattern rate of the DMD, we demonstrated 4.2 ms acquisition times for full spectra with a bandwidth of 450 nm (5250–4300 cm−1; 1.9–2.33 µm). Due to the programmability of the DMD, spectral regions of interest can be chosen freely, thus reducing acquisition times further, down to the sub-millisecond regime. The adjustable resolving power of the system accessed by means of computer simulations is discussed, quantified for different measurement modes, and verified by comparison with a state-of-the-art Fourier-transform infrared spectrometer. We show measurements of characteristic polymer absorption bands in different operation regimes of the spectrometer. The theoretical multiplex advantage of 8 was experimentally verified by comparison of the noise behavior of the spectral coding approach and a standard line-scan approach.

Speeding up eQTL scans in the BXD population using GPUs

The BXD family of mouse strains are an important reference population for systems biology and genetics that have been fully sequenced and deeply phenotyped. To facilitate interactive use of genotype-phenotype relations using many massive omics data sets for this and other segregating populations, we have developed new algorithms and code that enable near-real-time whole genome QTL scans for up to one million traits. By using easily parallelizable operations including matrix multiplication, vectorized operations, and element-wise operations, our method is more than 700 times faster than a R/qtl linear model genome scan using 16 threads. We used parallelization of different CPU threads as well as GPUs. We found that the speed advantage of GPUs is dependent on problem size and shape (the number of cases, number of genotypes, and number of traits). Our approach is ideal for interactive web services, such as GeneNetwork.org that need to display results in real-time. Our implementation is available as the Julia language package LiteQTL at https://github.com/senresearch/LiteQTL.jl.

Robust Guided Wave Tomography Method for Large and Irregular Defects

Finding a fast, robust way to quantitatively measuring the remaining wall thickness of complex structures when multiple defects exist is one of the leading challenges in Nondestructive Testing (NDT). Traditional inversion algorithms like ray tomography and full waveform inversion (FWI) suffered from problems like convergence, limited resolution and slow speed. Diffraction tomography (DT) has speed advantage over the preceding methods and its resolution can be further amplified by integrating with other methods like bent-ray tomography and iteration. However, DT can only detect shallow and small defects. Compared with those methods, convolutional neural network (CNN) opens a new way for quantitative defect imaging, as with pre-trained data it can achieve significant speed and resolution than the traditional methods. In this paper, we investigated the performance of CNN in imaging multiple defects and the inversion results show that when dealing with multiple defects with complex shape on a plate-like structure, CNN can achieve better resolution than other methods with maximum errors below 0.54mm in most regions. This research provides the experimental guidance for future study in finding the possible ways to improve the resolution of the algorithms.

Hybrid analytic and machine-learned baryonic property insertion into galactic dark matter haloes

While cosmological dark matter-only simulations relying solely on gravitational effects are comparably fast to compute, baryonic properties in simulated galaxies require complex hydrodynamic simulations that are computationally costly to run. We explore the merging of an extended version of the equilibrium model, an analytic formalism describing the evolution of the stellar, gas, and metal content of galaxies, into a machine learning framework. In doing so, we are able to recover more properties than the analytic formalism alone can provide, creating a high-speed hydrodynamic simulation emulator that populates galactic dark matter haloes in N-body simulations with baryonic properties. While there exists a trade-off between the reached accuracy and the speed advantage this approach offers, our results outperform an approach using only machine learning for a subset of baryonic properties. We demonstrate that this novel hybrid system enables the fast completion of dark matter-only information by mimicking the properties of a full hydrodynamic suite to a reasonable degree, and discuss the advantages and disadvantages of hybrid versus machine learning-only frameworks. In doing so, we offer an acceleration of commonly deployed simulations in cosmology.

Rapid neural encoding of the contrast between native and nonnative speech in the alpha band

More than half of the world's population is multilingual, yet it is not known how the human brain encodes the perception of native vs. nonnative speech. To find out, we asked German native speakers to detect the onset of native and nonnative (English and Turkish) vowels in a roving standard stimulation. Using EEG, we show that nonnativeness is robustly registered by an increase in phase coherence in the alpha band (8-12 Hz), beginning as early as ~100 ms after stimulus onset and lasting more than 200 ms. The alpha band effect is speech-specific, successfully predicts the response speed advantage of nonnative speech, and grants ~90% decoding accuracy in distinguishing native vs. nonnative speech irrespective of language familiarity. We propose alpha phase coherence as a candidate neural channel for the online resolution of the native-nonnative contrast in the adult brain.

Fast Fabrication of Fishnet Optical Metamaterial Based on Femtosecond Laser Induced Stress Break Technique

To speed up the fabrication of optical metamaterials by making use of the fast speed advantage of femtosecond laser preparation, a metamaterial appropriate for femtosecond laser processing was designed, and the interaction between femtosecond laser and metal-dielectric-metal fishnet stacks was investigated in detail. Two kinds of processing mechanisms, thermal melting and stress break, were revealed during the fabrication. The thermal melting process, dominated by the interaction of femtosecond laser with metals, makes the upper and lower metal layers adhere to each other, which leads to the magnetic resonance impossible. The stress break process, dominated by the interaction of femtosecond laser with dielectrics, can keep the upper and lower metal coatings isolated. Fishnet optical metamaterial was fabricated by femtosecond laser-induced stress break technique, using back side ablation, high numerical aperture and super-Gaussian beam. The resolution and speed can reach 500 nm, and 100 units/s, respectively. Spectrophotometer measurement results proved that the magnetic resonances were found in the fishnet nanostructure. The theoretical refractive index of the metamaterial on a glass substrate reached −0.12 at the wavelength of 3225 nm. It proved that femtosecond laser-induced stress break was a good and fast tool during the fabrication of optical metamaterials.

Level and speed of acquisition integration and their effects on technological performance

PurposeThis study concerns two aspects of the integration process critical for the success of acquisitions: (1) levels of human integration and task integration and (2) speeds of human integration and task integration. The purpose of this study is to examine the interaction effects of human/task integration level and human integration speed advantage on acquisition performance.Design/methodology/approachThis study collected data of companies in the Taiwanese high-tech industries at the financial, organizational and industrial levels to examine the proposed hypotheses. Corporate financial and patent data were collected from the Taiwan Securities and Futures Commission databases and the Intellectual Property Office (IPO) databases. The organizational level data were collected from 142 publicly traded related acquisitions from 2008 to 2009 in the Taiwanese high-tech industries.FindingsThe results show that (1) a high level of human integration positively affects technological performance; (2) the interaction term of human integration level and human integration speed advantage (i.e., relatively faster human integration coupled with slower task integration) positively affects technological performance; and (3) the interaction term of task integration level and human integration speed advantage positively affects technological performance.Originality/valueThe originality of this study lies in advancing our understanding of how complex interactions between human/task integration level and human integration speed advantage affect acquisition performance.