Combining deep learning and intelligent biometry to extract ultrasound standard planes and assess early gestational weeks.

OBJECTIVES: To develop and validate a fully automated AI system to extract standard planes, assess early gestational weeks, and compare the performance of the developed system to sonographers. METHODS: In this three-center retrospective study, 214 consecutive pregnant women that underwent transvaginal ultrasounds between January and December 2018 were selected. Their ultrasound videos were automatically split into 38,941 frames using a particular program. First, an optimal deep-learning classifier was selected to extract the standard planes with key anatomical structures from the ultrasound frames. Second, an optimal segmentation model was selected to outline gestational sacs. Third, novel biometry was used to measure, select the largest gestational sac in the same video, and assess gestational weeks automatically. Finally, an independent test set was used to compare the performance of the system with that of sonographers. The outcomes were analyzed using the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and mean similarity between two samples (mDice). RESULTS: The standard planes were extracted with an AUC of 0.975, a sensitivity of 0.961, and a specificity of 0.979. The gestational sacs' contours were segmented with a mDice of 0.974 (error less than 2 pixels). The comparison showed that the relative error of the tool in assessing gestational weeks was 12.44% and 6.92% lower and faster (min, 0.17 vs. 16.6 and 12.63) than that of the intermediate and senior sonographers, respectively. CONCLUSIONS: This proposed end-to-end tool allows automatic assessment of gestational weeks in early pregnancy and may reduce manual analysis time and measurement errors. CLINICAL RELEVANCE STATEMENT: The fully automated tool achieved high accuracy showing its potential to optimize the increasingly scarce resources of sonographers. Explainable predictions can assist in their confidence in assessing gestational weeks and provide a reliable basis for managing early pregnancy cases. KEY POINTS: • The end-to-end pipeline enabled automatic identification of the standard plane containing the gestational sac in an ultrasound video, as well as segmentation of the sac contour, automatic multi-angle measurements, and the selection of the sac with the largest mean internal diameter to calculate the early gestational week. • This fully automated tool combining deep learning and intelligent biometry may assist the sonographer in assessing the early gestational week, increasing accuracy and reducing the analyzing time, thereby reducing observer dependence.

A New Approach for Fatigue Reliability Analysis of Thin-Walled Structures with DC-ILSSVR.

Fatigue analysis is of great significance for thin-walled structures in the spacecraft industry to ensure their service reliability during operation. Due to the complex loadings of thin-walled structures under thermal-structural-acoustic coupling conditions, the calculation cost of finite element (FE) simulations is relatively expensive. To improve the computational efficiency of dynamic reliability analysis on thin-walled structures to within acceptable accuracy, a novel probabilistic approach named DC-ILSSVR was developed, in which the rotation matrix optimization (RMO) method was used to initially search for the model parameters of least squares support vector regression (LS-SVR). The distributed collaborative (DC) strategy was then introduced to enhance the efficiency of a component suffering from multiple failure modes. Moreover, a numerical example with respect to thin-walled structures was used to validate the proposed method. The results showed that RMO performed on LS-SVR model parameters provided competitive prediction accuracy, and hence the reliability analysis efficiency of thin-walled pipe was significantly improved.

Up-regulated ADP-Ribosylation factor 3 promotes breast cancer cell proliferation through the participation of FOXO1.

ADP-ribosylation factor 3 (ARF3) is a member of the KRAS proto-oncogene, GTPase(Ras) super-family of guanine nucleotide-binding proteins that mediates Golgi-related mitosis, but its role in malignant cells is unclear. In the present study, we found that mRNA and protein expression of ARF3 is up-regulated in breast cancer cells. Immunohistochemical analysis of 167 paraffin-embedded archived breast cancer tissues showed that ARF3 expression was localized primarily in the cytoplasm and was significantly up-regulated in malignant specimens compared to benign specimens. There were strong associations between ARF3 expression and clinicopathological characteristics in breast cancer. We also found that overexpressing ARF3 promoted, while silencing endogenous ARF3 inhibited, the proliferation of breast cancer cells by regulating cell cycle G1-S transition. Moreover, the pro-proliferative effect of ARF3 on breast cancer cells was associated with inactivation of the forkhead box O1 (FOXO1) transcription factor. ARF3 promotes breast cancer cell proliferation through the participation of FOXO1 and represents as a novel prognostic marker and therapeutic target for breast cancer.

Analysis of the clinical outcomes of fetal bowel dilatation combined with other abnormal ultrasonographic features.

OBJECTIVES: To explore the significance of fetal bowel dilatation combined with other abnormal ultrasound features in the diagnosis of gastrointestinal malformation. METHODS: A retrospective study of fetuses with bowel dilatation was performed, from August 2012 to October 2015. All the cases were identified from the ultrasound database and all observations of the relationship of prenatal abnormal abdominal ultrasound features and intestinal malformation were performed through the infancy stage. RESULTS: We found 52 fetuses with prenatal suspicion of bowel dilatation. Of these, 20 cases were surgically confirmed to have intestinal malformation, 13 cases had no abnormal bowel loops after birth, 8 cases had abnormal intestinal features while no surgical intervention was performed after birth, 10 cases were lost to follow-up and 1 fetus died in utero at 34 weeks of gestation. Forty cases with full data were divided into three groups, including Group A (Small bowel dilatation combined with other features vs. Isolated small bowel dilatation), Group B (Colonic bowel dilatation combined with other features vs. Isolated colonic bowel dilatation) and Group C (Bowel dilatation combined with other features vs. Isolated bowel dilatation). The intestinal malformation occurrence rates were 73.33% vs. 31.25% in Group A, 50% vs. 25% in Group B, and 70% vs. 30% in Group C. These results suggest that malformation occurs at a lesser frequency in simple bowel dilatation versus bowel dilatation in combination with other abnormal ultrasound features (p = .026), similarly in simple small bowel dilatation versus small bowel dilatation in combination with other abnormal ultrasound features (p = .032). CONCLUSIONS: Prenatal bowel dilatation in combination with other abnormal ultrasound features, especially small bowel dilatation in combination with other abnormal ultrasound features, detected in the second and third trimesters, tended to indicate intestinal malformation, which contributes to enhance the accuracy of prenatal diagnosis of intestinal malformation.

Independent risk factors for morphological abnormalities of the internal carotid artery.

PURPOSE: Whether carotid abnormalities (CAs) are congenital or acquired, and whether they are risk factors for cerebrovascular diseases or medical curiosities, is still under debate. The aim of this study was to identify the independent risk factors for CAs. METHODS: In this study, subjects (aged 3 days to 102 years) were divided into nine age groups, divided by 10-year intervals. The observed CAs were prospectively classified, according to the nature of the abnormality. Analyses of the factors associated with CAs were performed by univariate and multivariate logistic regression models. RESUITS: Of the 1,838 internal carotid arteries studied, 552 had CA (30%). There were peaks in the prevalence of CA in the 0-9-year-old group and in the above 80 years group; the lowest rate was found among the subjects in the 20-39-year-old group. The significant predictors of CA were age (OR 1.986, in the60-69-year-old group; OR 2.283, in the 70-79-year-old group; and OR 3.566, in the > 80-year-old group vs the 40-49-year-old group reference group), female gender (OR 1.799), hypertension (OR 1.52), and carotid artery atherosclerosis (OR3.176). CAs in youngersubjects were also associated with age (OR 9.894 in the 0-9-year-old group; and OR 3.294 in the 10-19-year-old group vs the 30-39-year-old group reference group). CONCLUSIONS: Both congenital and acquired factors may be involved in CAs. In subjects> or = 40years old, CAs seem to be more associated with acquired factors, including ageing, gender, hypertension, and atherosclerosis.

Objective differential characteristics of cystic biliary atresia and choledochal cysts in neonates and young infants: sonographic findings.

OBJECTIVES: To prospectively evaluate the objective differential characteristics between cystic biliary atresia and choledochal cysts on sonography among neonates and young infants. METHODS: Twenty-three patients who had sonographic findings of a portal cyst and a final diagnosis were included. Their final diagnoses were cystic biliary atresia in 12 patients and choledochal cysts in 11. All of them underwent detailed sonographic scanning. Data for cystic biliary atresia and choledochal cyst groups were compared by the χ(2) test for categorical variables and an unpaired t test for continuous variables. RESULTS: The triangular cord sign was detected in 11 patients in the atresia group but in none in the cyst group (P < .001). Nine of 11 patients in the cyst group had dilatation of intrahepatic bile ducts, whereas none in the atresia group had that feature (P < .001). Sonography also showed sludge deposits in the cysts in 6 of 11 patients in the cyst group, whereas none in the atresia group had sludge deposits (P = .005). The mean width and length of the cysts in the cyst group were significantly larger than those in the atresia group (P< .05 for both). The mean hepatic artery diameter was significantly larger in the atresia group than in the cyst group (P < .001). The difference in gallbladder abnormalities between the atresia (n = 11) and cyst (n = 0) groups was also significant (P < .001). When all specific sonographic features were used, all patients were correctly classified into the atresia and cyst groups. CONCLUSIONS: The triangular cord sign, intrahepatic bile duct dilatation, and echoic cysts might be regarded as objective sonographic features for differentiating cystic biliary atresia and choledochal cysts. Other sonographic features might be very supportive.