Method: In this study, we employed a pre-trained deep convolutional neural network (PCNN) to automatically segment lesions from ultrasound images. Specifically, our PCNN based method used whole images of normal tissues and lesions as inputs and then generated the segmentation probability maps as outputs. A pre-training strategy was used to improve the performance of the PCNN based model. Additionally, we compared the performance of our approach with that of the common convolutional neural network segmentation methods on the same dataset. 

Results: We validated on ultrasound images of thyroid nodules and breast nodules. The experimental results were shown in true positive rate (TP), false positive rate (FP), overlap metric (OM) and dice ratio (DR). Specifically, for thyroid nodule segmentation, our method could achieve an average of OM, DR, TP, FP as 0.8943, 0.9558, 0.9694, 0.0569 on overall folds, respectively. For breast nodule segmentation, our method could achieve an average of OM, DR, TP, FP as 0.8572, 0.9001, 0.9497, 0.8619, respectively. 

Conclusion: Our proposed method is fully automatic without any user interaction and may be good enough to replace the time-consuming and tedious manual segmentation approach, demonstrating the potential clinical applications.

Methods: TLCEUS was performed in 88 patients with newly diagnosed EBC. Methylene blue dye was percutaneously injected into enhanced sentinel lymph nodes (SLNs) under ultrasound guidance, followed by standard SLNB and axillary lymph node dissection. Enhancement patterns and the arriving time (AT) of contrast agent within SLNs were evaluated. Histopathological examination of dissected nodes was performed to confirm metastasis. 

Results: A total of 95 enhanced SLNs were identified and biopsied in 86 of 88 patients with identification rate of 97.7%. The specificity was 75.0%, sensitivity was 83.3%, and false-negative rate was 16.7%. Contrast-enhanced SLNs with type I, type II, and type III patterns had a metastatic positive rate of 11.4% (5/44), 57.1% (12/21) and 80.0% (24/30), respectively. Metastatic positive SLNs showed a mean AT of 61.6 ± 58.7 s while metastatic negative SLNs showed a mean AT of 41.3 ± 19.9 s, which was statistically significantly different. 

Conclusion: The TLCEUS/blue dye method can be used as an alternative to the radioisotope/blue dye method for its feasibility and accuracy.

Methods: A total of 152 pregnant women who received prenatal examination in Beijing Obstetrics and Gynecology Hospital during January and October 2017. On the basis of the clinical diagnostic criteria for preterm birth, all of pregnant women were divided into two groups: There are 50 patients with the diagnosis of threatened premature or preterm labor in the study group and 102 patients without diagnosis of threatened premature or preterm labor in the control group. Measure cervical length by transvaginal ultrasound and obtain three cervical elasticity indexes by strain elastography: Closed internal cervical os strain rate, external cervical os strain rate and midpoint of the cervical canal strain rate. In the study group the pregnant women were divided into three teams: Length of cervix is 2 cm or less (10 cases) for T1, 2 to 3 cm (21 cases) for T2 and 3 cm or more (19 cases) for T3. Independent sample t-test was used to analyze the difference between the two groups. LSD-t test was used in the two groups. The receiver operating characteristic curve was used to analysis these four indicators. Statistical significance was defined as a p value < 0.05. 

Results: CL was lower and CIS was higher in the study group (2.67 ± 0.76 vs. 3.85 ± 0.65, p < 0.05; 0.53 ± 0.35 vs. 0.27 ± 0.24, p < 0.05). There was a significant difference with CIS in different cervical length team (p = 0.017) and the mean value of CIS in T1 was higher than the last two teams. The cutoff value of CL maximizing the accuracy of diagnosis was 3.2 cm; area under the curves (AUC), sensitivity, specificity, accuracy, positive predictive value (PPV) and negative predictive value (NPV) were 0.877, 73.3%, 87.0%, 82.2%, 73.5%, 86.4% respectively. Combining CL and CIS, the results of AUC, sensitivity, specificity, accuracy, PPV and NPV (0.891, 73.3%, 91.3%, 84.9%, 80.0%, 86.9%) increased in predicting the risk of spontaneous preterm birth , and the cutoff values of the CL and CIS were 3.1 cm and 0.35%. 

Conclusion: The shortening of CL and the increasing CIS are closely related to the occurrence of premature birth and can be used as the indexes to predict the preterm labor. CIS can be used as an indicator to monitor early preterm changes. The combination of CL and CIS may have a role in assessing the risk of spontaneous preterm birth.

Methods: A retrospective analysis, 66 breast cancer patients with 101 ipsilateral axillary SLNs checking by CEUS, was confirmed by pathology. SLCs and SLNs positions were marked and then the rough needle biopsy pathology was used to confirm the benign or malignant SLNs. According to the different perfusion, SLNs were divided into five groups and evaluated by scores. MecNemar test and receiver operating characteristic (ROC) curve were used to analyze the statistical analysis. The difference of p < 0.05 was statistically significant. The cut-off value was finally determined, and the results of ultrasound screening test were evaluated. 

Results: A total of 66 patients with 101 SLNs were successfully evaluated, including 42 cases of metastatic lymph nodes, and 59 cases of benign lymph nodes. The area under the ROC curve was 0.752 (p < 0.05). To ROC curve, the cut off value was 3.5. The scores 4 and 5 of lymph nodes were negative, and scores 1, 2, 3 of lymph nodes were positive. The sensitivity, specificity, false negative rate, false positive rate, and the coincidence rate of the screening test were 44.19%, 93.10%, 55.81%, 6.9%, 72.28% respectively. The value of positive predictive and the negative predictive were 82.61%, 69.23% respectively. 

Conclusion: To evaluate SLCs shape and judge benign and malignant SLNs, percutaneous CEUS has a high practical value.

Methods. Eighty patients who came to Shenyang Children Hospital with vaginal discharge, vaginal bleeding, vulva redness and swelling were evaluated with transperineal ultrasonography and four typically cases were chosed which have tumor, vagina foreign body, vaginitis and imperforate hymen. Color Doppler ultrasonography was applied to these abnormities. 

Results When transperineal ultrasonography was applied to these abnormities, the tumor had the sigh of occupying lesion and sparse color flow. Vagina foreign body showed high-echo strip object or moniliform in vagina accompany with vaginal dropsy. A 15 days neonate girl with hydrocolpocele was diagnosed imperforate hymen later. 

Conclusion Transperineal ultrasonography can be used to enhance the diagnosis of vagina illness in young girls.

Methods: The ADDH was evaluated by sonography in both study subjects (CHD group, n = 506) and age-matched normal controls (normal group, n = 506). The CHD group was divided into different subgroups according to evident anomalies in the cardiovascular system. ADDH differences between CHD subgroups and their controls were calculated with a paired Student's t-test. Possible sex differences were determined by an unpaired Student's t-test on the whole group. 

Results: There were no significant differences in the ADDH between all CHD subgroups and age-matched controls (p > 0.05). No difference was found in the ADDH between female and male fetuses (p > 0.05). 

Conclusion: The ADDH in fetuses with CHD is similar to that in controls. Hence, ratios between size of involved cardiovascular structures and ADDH may be a useful parameter for monitoring fetuses with CHD.

Methods: A total of 30 patients (29 men and 1 woman; a mean age of 35.4 years) with Kimura disease confirmed by surgery and pathology were enrolled in the study. Ultrasonographic findings were analyzed retrospectively and compared with clinical features and laboratory examinations to better understand the relationship among them. 

Results: All patients presented with painless, progressively enlarged soft tissue masses in head and neck, axillary, inguinal regions or limb extremities. Laboratory results showed elevation of eosinophils in both peripheral blood and serum IgE in 12 of 30 cases. Ultrasound findings of the lesions in Kimura disease were divided into three types: Type I: characterized by simple lymph node enlargement; Type II: characterized by flake-like thickening of the lesion without mass effect, and enlargement of nearby lymph nodes; Type III: a tumor-like lesion associate with enlargement of nearby lymph nodes. 

Conclusion: The Kimura disease has characteristic ultrasonographic characteristics of lymph node and adjacent structures. When combined with patient’s clinical information and laboratory tests, ultrasound examination can provide useful information for making clinical diagnosis and management.

Methods: A retrospective analysis was conducted for 21 pathologically confirmed paediatric cases of CPAM by reviewing the ultrasonography images and comparing them with the clinical and pathologic results. 

Results: All 21 paediatric cases of CPAM had lesions in unilateral lobes, with five cases involving the left lung and 16 cases involving the right lung. There were six cases with type 1 ultrasonography manifestations, 12 cases with type 2 ultrasonography manifestations and one case with a type 4 ultrasonography manifestation. One case was misdiagnosed as pulmonary sequestration (PS) by ultrasonography, and another case had no abnormalities based on ultrasonography. With respect to the pathologic results, six type 1 (28.6%), 13 type 2 (61.9%) and one type 4 (4.8%) were found, and one type 2 was complicated with PS (4.8%). The ultrasonography diagnoses agreed with surgical pathology reports in 18 cases (85.7%). 

Conclusion: Ultrasonography is feasible and accurate in identifying subtype within postnatally diagnosed pediatric CPAM lesions.

Methods: Electronic databases of PubMed, Embase, Web of Science, Scopus were searched up to June 2018 for studies comparing the standardized mean differences (SMD) of size, amylase and carbohydrate antigen 19-9 (CA199) levels between pre- and post-operation for patients with pancreatic cancer. Sensitivity and stratified analyses were conducted. Quality was estimated using Newcastle-Ottawa Scale (NOS). 

Results: We finally identified 10 studies including 203 participants during a mean 7.06 months of follow-up (range 1 to 29 months). The meta-analyses showed the declined tumor size at 6 months post-IRE but unchanged at 1 month, and increased amylase level at 1-day post-IRE while unchanged at the 1 week. No significant difference of CA199 level was observed between pre-IRE and post-IRE at 1 week and 1 month. No risk of publication bias was detectable, and the favorable quality and validity of all outcomes were assessed based on NOS. 

Conclusions: IRE may be a relatively state-of-the-art therapy option for most patients with LAPC if imaging or explorative laparotomy indicated that LAPC was unable to be successfully resected.