Search

Background: Fetal brain magnetic resonance imaging (MRI)‐based gestational age prediction has been widely used to characterize normal fetal brain development and diagnose congenital brain malformations. Purpose: The uncertainty of fetal position and external interference leads to variable localization and direction of the fetal brain. In addition, pregnant women typically concentrate on receiving MRI scans during the fetal anomaly scanning week, leading to an imbalanced distribution of fetal brain MRI data. The above‐mentioned problems pose great challenges for deep learning‐based fetal brain MRI gestational age prediction. Methods: In this study, a pyramid squeeze attention (PSA)‐guided dynamic feature fusion CNN (PDFF‐CNN) is proposed to robustly predict gestational ages from fetal brain MRI images on an imbalanced dataset. PDFF‐CNN contains four components: transformation module, feature extraction module, dynamic feature fusion module, and balanced mean square error (MSE) loss. The transformation and feature extraction modules are employed by using the PSA to learn multiscale and multi‐orientation feature representations in a parallel weight‐sharing Siamese network. The dynamic feature fusion module automatically learns the weights of feature vectors generated in the feature extraction module to dynamically fuse multiscale and multi‐orientation brain sulci and gyri features. Considering the fact of the imbalanced dataset, the balanced MSE loss is used to mitigate the negative impact of imbalanced data distribution on gestational age prediction performance. Results: Evaluated on an imbalanced fetal brain MRI dataset of 1327 routine clinical T2‐weighted MRI images from 157 subjects, PDFF‐CNN achieved promising gestational age prediction performance with an overall mean absolute error of 0.848 weeks and an R2$R^2$ of 0.904. Furthermore, the attention activation maps of PDFF‐CNN were derived, which revealed regional features that contributed to gestational age prediction at each gestational stage. Conclusions: These results suggest that the proposed PDFF‐CNN might have broad clinical applicability in guiding treatment interventions and delivery planning, which has the potential to be helpful with prenatal diagnosis. [ABSTRACT FROM AUTHOR]

Voltage source converter‐high‐voltage direct current (VSC‐HVDC) is an effective way to ensure the reliable supply of offshore island power. However, it is easily liable to direct current (DC) oscillation or even instability due to the interaction between the rectifier and inverter stations. In order to solve this problem, this paper analyzes the stability of the VSC‐HVDC system caused by DC capacitors based on VSC impedance modeling and impedance stability criterion. Besides, an adaptive virtual impedance control strategy to suppress the DC oscillation is proposed. The impedance models of the VSC in the rectifier and inverter stations are constructed and verified. Then, the influence of the DC capacitors on the system stability is analyzed based on impedance stability criteria. The analysis shows that the decrease of DC capacitors will cause DC oscillation, and the DC virtual impedance is used to damping the oscillation. Finally, this paper presents a simple but effective adaptive fuzzy segmented virtual impedance control strategy to suppress the oscillation. Electromagnetic transient simulation and hardware‐in‐the‐loop (HIL) test results show that the proposed adaptive virtual impedance is more effective and has better robustness than the conventional virtual impedance. © 2018 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc. [ABSTRACT FROM AUTHOR]

Soybean cyst nematode (SCN), Heterodera glycines , poses a significant threat to global soybean production. Heilongjiang, the largest soybean-producing province in China, contributes over 40% to the country's total yield. This province has much longer history of SCN infestation. To assess the current situation in Heilongjiang, we conducted a survey to determine the SCN population density and virulence phenotypes during 2021-2022 and compared the data with a previous study in 2015. A total of 377 soil samples from 48 counties representing eleven major soybean-planting regions were collected. The prevalence of SCN increased from 55.4% in 2015 to 59% in the current survey. The population densities ranged from 80 to 26,700 eggs and juveniles per 100 cm3 of soil. Virulence phenotypes were evaluated for 60 representative SCN populations using the HG type test, revealing nine different HG types. The most common virulence phenotypes were HG types 7 and 0, accounting for 56.7% and 20% of all SCN populations, respectively. The prevalence of populations with a reproductive index (FI) greater than 10% on PI548316 increased from 64.5% in 2015 to 71.7%. However, the FI on the commonly used resistance sources PI 548402 (Peking) and PI 437654 remained low at 3.3%. These findings highlight the increasing prevalence and changing virulence phenotypes of SCN in Heilongjiang. They also emphasize the importance of rotating soybean varieties with different resistance sources and urgently identifying new sources of resistance to combat SCN.