Ultrasonographic Segmentation of Fetal Lung with Deep Learning

The morbidity and mortality of the fetus is related closely with the neonatal respiratory morbidity, which was caused by the immaturity of the fetal lung primarily. The amniocentesis has been used in clinics to evaluate the maturity of the fetal lung, which is invasive, expensive and time-consuming. Ultrasonography has been developed to examine the fetal lung quantitatively in the past decades as a non-invasive method. However, the contour of the fetal lung required by existing studies was delineated in manual. An automated segmentation approach could not only improve the objectiveness of those studies, but also offer a quantitative way to monitor the development of the fetal lung in terms of morphological parameters based on the segmentation. In view of this, we proposed a deep learning model for automated fetal lung segmentation and measurement. The model was constructed based on the U-Net. It was trained by 3500 data sets augmented from 250 ultrasound images with both the fetal lung and heart manually delineated, and then tested on 50 ultrasound data sets. With the proposed method, the fetal lung and cardiac area were automatically segmented with the accuracy, average IoU, sensitivity and precision being 0.98, 0.79, 0.881 and 0.886, respectively.

The soil configuration on granite residuals affects Benggang erosion by altering the soil water regime on the slope

A permanent collapsing gully,locally called Benggang,formed on slopes with deep granite red soil and is a type of unique gully erosion widely prevalent in southern China.Three different soil configurations(SC),ie,red-transition-sandy(SC I,the transition is the soil layer between the red soil and the sandy soil layer),transition-sandy(SC II)or sandy(SC III)are usually present in the soil profile of the Benggang slope.However,little attention has been paid to impacts of SCs on the triggering of Benggang erosion.In this study,we aimed to explore the relationships between soil water content(SWC)and triggering of Benggang erosion under different SC conditions.The soil properties of different soil layers were measured and the SWC at depths of 20,40,60,and 80 cm were monitored at 5-min intervals along a typical Benggang(SC I)during 2016-2018.The SWC of Benggang slopes with different SCs were simulated by VADOSE/W model.Results showed that the red soil layer had a higher water retention capacity and shear strength than the sandy soil layer.Even if the SWC is higher(e.g.,0.42 cm^(3)/cm^(3))at red soil layer or transition layer,the corresponding shear strength is greater than that of sandy soil layer with a lower SWC(e.g.,032 cm^(3)/cm^(3)).Relationships between shear strength and SWC of different soil layers indicate that Benggang erosion is triggered by an increase in the SWC in the deep sandy layer.Results also showed that differences exist in the SWC distribution among the different SCs.The SWC is higher in topsoil than in deeper soil in SC I and SC II,while in SC III,the opposite trend is observed.These results revealed that the presence of the red soil or transition layer can reduce the infiltration of rainwater into the deep sandy layer,thus can reduce the possibility of collapse.Our results show that the SC affects the stability of the headwall,and results provide great significances to guide the mitigation of Benggang erosion.

Effect of joint structure and slope direction on the development of collapsing gully in tuffaceous sandstone area in South China

This study focuses on the collapsing gullies in tuffaceous sandstone area and investigates the slope di-rection and morphological characteristics of the main and branch gullies. Furthermore, we assess the structural characteristics of the rock joints within this area, including their strike, dip direction and dip angle. The results show that there are 405 collapsing gullies in the study area. The slope directions associated with collapsing gullies and the directions of the main gullies largely fall within the ranges of NE20°-NE90°, SE90°-SE160°, SW240°-SW270°, and NW270°-NW290°. The collapsing gullies include 1103 branch gullies in total, most of which have directions that fall within the ranges of NE20°-NE40°, NE50°-NE70°, NW280°-NW300°, and NW330°-NW350°. The joints in the bedrock are directional and regional, and they can be divided into two main groups. The number of southward dip directions is greater than the number of northward dip directions, and most of the measured dip angles are greater than 60°. The mean dip angle is greatest for joints with measured strike values of NW280°-NW290°, with a value of 85.2°. The development of collapse gullies is affected by both the slope direction and joints. The slope direction determines the direction of the main gullies, with a correlation coefficient of 0.809 (P<0.01). The branch gullies are mainly affected by joints, with a correlation coefficient of 0.876 (P<0.01). The joint structure also influences the degree of development of the collapsing gullies, and the average depth of the gullies that parallel the dominant joint orientation is significantly larger than that of gullies with other directions. Moreover, the average depth of the gullies associated with the dip angle of 85.2° measured relative to the joint strike is 6.89 m, which is significantly greater than that associated with lower dip angles. The dip angles of joints have an important effect on the infiltration of water, and high dip angles accelerate the erosion associated with collapsing gullies.