ABSTRACT
BACKGROUND@#Prenatal evaluation of fetal lung maturity (FLM) is a challenge, and an effective non-invasive method for prenatal assessment of FLM is needed. The study aimed to establish a normal fetal lung gestational age (GA) grading model based on deep learning (DL) algorithms, validate the effectiveness of the model, and explore the potential value of DL algorithms in assessing FLM.@*METHODS@#A total of 7013 ultrasound images obtained from 1023 normal pregnancies between 20 and 41 + 6 weeks were analyzed in this study. There were no pregnancy-related complications that affected fetal lung development, and all infants were born without neonatal respiratory diseases. The images were divided into three classes based on the gestational week: class I: 20 to 29 + 6 weeks, class II: 30 to 36 + 6 weeks, and class III: 37 to 41 + 6 weeks. There were 3323, 2142, and 1548 images in each class, respectively. First, we performed a pre-processing algorithm to remove irrelevant information from each image. Then, a convolutional neural network was designed to identify different categories of fetal lung ultrasound images. Finally, we used ten-fold cross-validation to validate the performance of our model. This new machine learning algorithm automatically extracted and classified lung ultrasound image information related to GA. This was used to establish a grading model. The performance of the grading model was assessed using accuracy, sensitivity, specificity, and receiver operating characteristic curves.@*RESULTS@#A normal fetal lung GA grading model was established and validated. The sensitivity of each class in the independent test set was 91.7%, 69.8%, and 86.4%, respectively. The specificity of each class in the independent test set was 76.8%, 90.0%, and 83.1%, respectively. The total accuracy was 83.8%. The area under the curve (AUC) of each class was 0.982, 0.907, and 0.960, respectively. The micro-average AUC was 0.957, and the macro-average AUC was 0.949.@*CONCLUSIONS@#The normal fetal lung GA grading model could accurately identify ultrasound images of the fetal lung at different GAs, which can be used to identify cases of abnormal lung development due to gestational diseases and evaluate lung maturity after antenatal corticosteroid therapy. The results indicate that DL algorithms can be used as a non-invasive method to predict FLM.
Subject(s)
Female , Humans , Infant , Infant, Newborn , Pregnancy , Algorithms , Deep Learning , Gestational Age , Lung/diagnostic imaging , Neural Networks, ComputerABSTRACT
<p><b>OBJECTIVE</b>To investigate the effects of artesunate (ART) on proliferation, cell cycle and apoptosis of SKM-1 cells in vitro and to explore the underlying mechanisms.</p><p><b>METHODS</b>After SKM-1 cells were treated with different concentrations of ART, the cell proliferation was determined by CCK-8 method. Apoptosis and distribution of cell cycle were detected by flow cytometry. Both DCFH-DA fluorescent probe and Fluo-3-Am fluorescent probe were used to detect the changes of intracellular reactive oxygen species (ROS) and calcium ion concentration. Western blot was used to measure the protein levels of BCL-2, BAX, BAD, P-BAD, survivin and XIAP.</p><p><b>RESULTS</b>ART obviously inhibited the growth of SKM-1 cells in time and dose-dependent manners (r = -0.841; r = 0.-786). The antioxidant trolox-pretreatment significantly decreased the growth inhibition effect of ART on SKM-1 cells. Caspase inhibitor Ac-DEVD-CHO partially reduced the growth inhibition effect of ART on SKM-1 cells. After treatment with ART for 24 hours, the apoptosis of SKM-1 cells was found, the cell cycle of SKM-1 was arrested in G0/G1 phase, ART could elevate the levels of calciumion and reactive orygen. ART could significantly down-regulate the protein expression levels of P-BAD and survivin in SKM-1 cells, and showed a highly negative correlation with ART dose (r = -0.909; r = -0.849). On the contrary, ART had no significant effect on expression levels of BAD and XIAP in SKM-1 cells, and after ART treatment, although BCL-2 protein expression was not significantly different when compared with control group, but the BCL-2/BAX ratio significantly decreased and highly negatively correlated with ART dose (r = -0.866).</p><p><b>CONCLUSION</b>The ART significantly suppresses the cell proliferation, induces the apoptosis and promoted cell cycle arrest at G0/G1 phase in SKM-1 cells. The mechanisms of ART anti-MDS is associated with the increase of intracellular calciumion concentration and ROS levels. In addition, the pro-apoptotic activity of ART may be involved in the regulation of BCL-2 /BAX ratio and the expressions of P-bad and survivin.</p>