Background Chest computed tomography (CT) is crucial for the detection of lung cancer, and many automated CT evaluation methods have been proposed. Due to the divergent software dependencies of the reported approaches, the developed methods are rarely compared or reproduced. Objective The goal of the research was to generate reproducible machine learning modules for lung cancer detection and compare the approaches and performances of the award-winning algorithms developed in the Kaggle Data Science Bowl. Methods We obtained the source codes of all award-winning solutions of the Kaggle Data Science Bowl Challenge, where participants developed automated CT evaluation methods to detect lung cancer (training set n=1397, public test set n=198, final test set n=506). The performance of the algorithms was evaluated by the log-loss function, and the Spearman correlation coefficient of the performance in the public and final test sets was computed. Results Most solutions implemented distinct image preprocessing, segmentation, and classification modules. Variants of U-Net, VGGNet, and residual net were commonly used in nodule segmentation, and transfer learning was used in most of the classification algorithms. Substantial performance variations in the public and final test sets were observed (Spearman correlation coefficient = .39 among the top 10 teams). To ensure the reproducibility of results, we generated a Docker container for each of the top solutions. Conclusions We compared the award-winning algorithms for lung cancer detection and generated reproducible Docker images for the top solutions. Although convolutional neural networks achieved decent accuracy, there is plenty of room for improvement regarding model generalizability.

Ground motion with strong-velocity pulses can cause significant damage to buildings and structures at certain periods; hence, knowing the period and velocity amplitude of such pulses is critical for earthquake structural engineering. However, the physical factors relating the scaling of pulse periods with magnitude are poorly understood. In this study, we investigate moderate but damaging earthquakes (Mw 6–7) and characterize ground-motion pulses using the method of Shahi and Baker (2014) while considering the potential static-offset effects. We confirm that the within-event variability of the pulses is large. The identified pulses in this study are mostly from strike-slip-like earthquakes. We further perform simulations using the frequency–wavenumber algorithm to investigate the causes of the variability of the pulse periods within and between events for moderate strike-slip earthquakes. We test the effect of fault dips, and the impact of the asperity locations and sizes. The simulations reveal that the asperity properties have a high impact on the pulse periods and amplitudes at nearby stations. Our results emphasize the importance of asperity characteristics, in addition to earthquake magnitudes for the occurrence and properties of pulses produced by the forward directivity effect. We finally quantify and discuss within- and between-event variabilities of pulse properties at short distances.

<p>Ground motions with strong pulses often bring significant damage to structures. The period and the amplitude of the strong-velocity pulses are critical for structural engineering and seismic hazard assessment. The scaling of pulses periods with magnitudes and the within-event variability of pulses is however poorly understood. In this study, we analyze two moderate earthquakes, namely 2016 Meinong earthquake and 2018 Hualien earthquake, using Shahi and Baker’s criteria (2014) to detect pulses. The observations in this study show that the amplitudes of the pulse decay with the distance from the source to the stations, and is also associated with the rupture direction from the asperity instead of the direction from the hypocenter. In addition, we further perform simulations using a simple FK method to clarify the causes of the variability of the pulse periods within and between events. We test the effect of faults dipping angles and the impacts of the asperity location and size. Through our simulations, we reveal that the amplitudes of the pulses in the shallow dipping fault are larger on the hanging wall than on the foot wall, and that the asperity properties has a large impact on the pulses periods and the amplitudes at the nearby stations. The results show that the asperity characteristics are critical for the occurrence of the strong-velocity pulses. The complete understanding of the kinematics of the rupture is then important for clarifying the effects of the strong-velocity pulses and improving ground-motions predictions.</p>

<p>A spectral decomposition of the Fourier amplitude spectra is applied to determine the source parameters of earthquakes (source spectral shape, stress drop) that have occurred in central-southern Europe. About 52 million waveforms recorded in the target area since the late ‘90s have been downloaded from the European Integrated Data Archive (EIDA) within the tool stream2segment (Zaccarelli et al., 2019), by using the event catalog of the International Seismic Centre (ISC) and innovative data quality assessment. A non-parametric decomposition approach in this study introduced a regionalization for the attenuation models into two spatial domains (southern and “active” Europe, northern and “stable” Europe). For each domain, a spectral attenuation with hypocentral distance model is simultaneously determined and used to remove regional specific propagation effects from the spectra of recordings. Once isolated from local site effects, the obtained source spectra of 4380 earthquakes of magnitude larger than 2.5 are fitted to a standard -model to determine the seismic moment, corner frequency and Brune stress drop. The scaling relationship, spatial variation and variability of these source parameters are finally derived and discussed.</p>