OBJECTIVES: To construct an intelligent analysis model for classifying fetal orientation during intrapartum ultrasound videos based on multi-feature fusion. METHODS: The proposed model consists of the Input, Backbone Network and Classification Head modules. The Input module carries out data augmentation to improve the sample quality and generalization ability of the model. The Backbone Network was responsible for feature extraction based on Yolov8 combined with CBAM, ECA, PSA attention mechanism and AIFI feature interaction module. The Classification Head consists of a convolutional layer and a softmax function to output the final probability value of each class. The images of the key structures (the eyes, face, head, thalamus, and spine) were annotated with frames by physicians for model training to improve the classification accuracy of the anterior occipital, posterior occipital, and transverse occipital orientations. RESULTS: The experimental results showed that the proposed model had excellent performance in the tire orientation classification task with the classification accuracy reaching 0.984, an area under the PR curve (average accuracy) of 0.993, and area under the ROC curve of 0.984, and a kappa consistency test score of 0.974. The prediction results by the deep learning model were highly consistent with the actual classification results. CONCLUSIONS The multi-feature fusion model proposed in this study can efficiently and accurately classify fetal orientation in intrapartum ultrasound videos.
Humans ; Female ; Ultrasonography, Prenatal/methods* ; Pregnancy ; Fetus/diagnostic imaging* ; Neural Networks, Computer ; Video Recording

Objective To explore the value of transvaginal ultrasound five-view method for diagnosing fetal congenital heart disease(CHD)in early pregnancy of high-risk pregnancy.Methods Totally 428 singleton fetuses at 11-13+6 weeks of gestation in high-risk pregnancies were prospectively enrolled.Fetal cardiac examinations were performed using two-dimensional grey scale imaging combined with CDFI via transabdominal ultrasound method,transvaginal ultrasound five-view method(upper abdominal transverse view,four-chamber view,three-vessel and trachea[3VT]view,ventricular outflow tract view and bilateral subclavian artery view)and transvaginal ultrasound two-view method(four-chamber view and 3VT view).Taken results of pregnant metaphase or late pregnancy fetal echocardiography,post-abortion pathology or postnatal neonatal echocardiography as gold standards,the diagnostic performances of the above 3 methods for diagnosing fetal CHD in early pregnancy were compared.Results CHD was diagnosed in 120(120/428,28.04％)fetuses.Transvaginal ultrasound five-view method made 11 false positives and 19 false negatives,with sensitivity,specificity,positive predictive value and negative predictive values for diagnosing fetal CHD in the first trimester of high-risk pregnancy of 84.17％,96.43％,90.18％and 93.99％,respectively,which were all significantly higher than of transvaginal ultrasound two-view method(70.83％,91.88％,77.27％and 88.99％)and transabdominal ultrasound method(72.50％,91.23％,76.32％and 89.49％;all P＜0.05).Conclusion Transvaginal ultrasound five-view method could be used to effectively diagnose fetal CHD in the first trimester of high-risk pregnancy.

Objective To explore the value of transvaginal ultrasound five-view method for diagnosing fetal congenital heart disease(CHD)in early pregnancy of high-risk pregnancy.Methods Totally 428 singleton fetuses at 11-13+6 weeks of gestation in high-risk pregnancies were prospectively enrolled.Fetal cardiac examinations were performed using two-dimensional grey scale imaging combined with CDFI via transabdominal ultrasound method,transvaginal ultrasound five-view method(upper abdominal transverse view,four-chamber view,three-vessel and trachea[3VT]view,ventricular outflow tract view and bilateral subclavian artery view)and transvaginal ultrasound two-view method(four-chamber view and 3VT view).Taken results of pregnant metaphase or late pregnancy fetal echocardiography,post-abortion pathology or postnatal neonatal echocardiography as gold standards,the diagnostic performances of the above 3 methods for diagnosing fetal CHD in early pregnancy were compared.Results CHD was diagnosed in 120(120/428,28.04％)fetuses.Transvaginal ultrasound five-view method made 11 false positives and 19 false negatives,with sensitivity,specificity,positive predictive value and negative predictive values for diagnosing fetal CHD in the first trimester of high-risk pregnancy of 84.17％,96.43％,90.18％and 93.99％,respectively,which were all significantly higher than of transvaginal ultrasound two-view method(70.83％,91.88％,77.27％and 88.99％)and transabdominal ultrasound method(72.50％,91.23％,76.32％and 89.49％;all P＜0.05).Conclusion Transvaginal ultrasound five-view method could be used to effectively diagnose fetal CHD in the first trimester of high-risk pregnancy.

Objective To observe the value of three-dimensional ultrasound virtual organ computer-aided analysis(VOCAL)technique for measuring choroid plexus volume(CPV)and lateral ventricle volume(LVV)in normal fetus of 11-13+6 gestational weeks.Methods Three-dimensional ultrasonic data of 428 normal singletons at 11-13+6 gestational weeks were retrospectively analyzed.The fetuses were divided into 11-11+6 weeks group(n=77),12-12+6 weeks group(n=225)and 13-13+6 weeks group(n=126).Fetal CPV,LVV and CPV/LVV ratio were measured and calculated using VOCAL technique and compared between groups.Pearson correlation analysis was performed to observe the correlations of CPV,LVV,CPV/LVV ratio with crown-rump length(CRL).Results Significant differences of CPV,LVV and CPV/LVV ratio were found among groups(all P＜0.05).Among 11-11+6 weeks group,12-12+6 weeks group and 13-13+6 weeks group,CPV and LVV increased progressively,while CPV/LVV ratio decreased progressively(all adjusted P＜0.05).During 11-13+6 gestational weeks,CPV and LVV were both highly positively correlated with CRL(r=0.904,0.929,both P＜0.05),while CPV/LVV ratio was highly negatively correlated with CRL(r=-0.893,P＜0.05).Conclusion Three-dimensional ultrasound VOCAL technique was feasible for measuring CPV and LVV in normal fetus of 11-13+6 gestational weeks.

Objective:To explore the application value of renal artery resistance index in the diagnosist of intra-abdominal hypertension (IAH) in critically ill patients.Methods:89 patients with risk factors of IAH in the intensive care unit of Zhangzhou Hospital Affiliated to Fujian Medical University from February 2022 to June 2022 were retrospective analyzed. The intra-abdominal pressure (IAP) were measured by bladder, and patients divided into IAH group (IAP≥12 mmHg) and non-IAH group (IAP <12 mmHg). The resistance index (RI) of the right renal aorta, segment artery and interlobar artery were measured by color Doppler ultrasound. The difference between the measurement indicators of patients in the IAH and non-IAH groups and the correlation with IAP were analyzed. As for different indicators to predict the diagnostic efficacy of IAH, ROC curve analysis was used to evaluate the effect. And further multivariate logistic regression analysis was to find independent risk predictor.Results:A total of 89 patients were included in the study, including 44 patients with normal IAP and 45 patients diagnosed IAH. There were significant differences in the right renal aorta, segmental artery and interlobar artery RI (all P<0.01). The interlobar artery RI≥0.698 was the highest diagnostic cut-off,area under the curve was 0.914, sensitivity was 82.2%, specificity was 97.7%, and Jordon index was 0.799. The Spearman correlation analysis of IAP and the ultrasonographic measurements revealed a strong correlation between right renal interlobar artery RI and IAP ( r=0.741, P<0.01). The multivariate logistic regression analysis showed that the right renal interlobar artery RI ( OR= 24.472, 95% CI:5.122~116.919, P<0.01) was an independent risk predictor of IAH ( P<0.01). Conclusion:Right renal interlobar artery RI had better diagnostic efficacy of IAH,renal ultrasound could be used as an alternative, non-invasive technique for the diagnosis and follow-up of IAH.

Objective:This study examines the application of echocardiography in the prenatal diagnosis of copy number variation (CNV) associated with fetal congenital heart disease (CHD).Methods:A retrospective analysis was conducted on 447 singleton pregnancies from Quanzhou Maternal and Child Care Hospital (Quanzhou Children's Hospital) from January 2019 to August 2022. These individuals underwent echocardiographic assessments suggestive of fetal CHD and subsequently received invasive prenatal diagnoses. Comprehensive karyotype analysis and chromosome microarray analysis (CMA) were performed for each case. The discrepancies in the chromosomal abnormality detection were analyzed between the results produced by CMA and karyotype analysis. Furthermore, differences in the detection of pathogenic copy number variation (pCNV) between the two methods in CHD cases with diverse cardiac phenotypes, including the presence or absence of extracardiac structural malformations, the type, and quantity of cardiac structural anomalies, were explored. Statistical analysis was conducted using the Chi-square test. Results:Compared with conventional karyotype analysis, CMA demonstrated a higher detection rate of fetal chromosomal abnormalities [10.5% (47/447) vs. 20.6% (92/447), χ 2=161.56, P<0.001]. In terms of distinct cardiac phenotypes, CHD cases with extracardiac structural anomalies displayed an escalated pCNV detection rate in comparison to isolated CHD cases [11.4% (45/394) vs. 32.1% (17/53), χ 2=16.68, P<0.001]. Within the cardiac structural anomaly subgroups, increased pCNV detection rates were observed in the septal defect subgroup, conotruncal malformation subgroup, and left ventricular malformation subgroup [18.4%(29/158), 25.9%(7/27), and 25.0%(7/28) vs. 7.6%(16/210); χ 2=9.15, 9.68, and 8.55, respectively, all P<0.05]. The CMA-identified pCNV correlated with CHD included 22q11.2 deletions/duplications in eight cases, 4p16.3 deletions in two cases, 11q23.3 microduplications in two cases, 1q21.1 microdeletions/microduplications in two cases, 4q28.3 microduplications in one case, and 10p15.3 microdeletions in one case. Conclusions:CMA technology exhibited an enhanced ability to detect pCNV in fetuses with CHD. Echocardiography can guide targeted CMA screening, thereby facilitating prenatal genetic assessment of CHD.

Objective:To explore the impact of benign prostatic hyperplasia (BPH)/lower urinary tract symptoms (LUTS) on the risk of cardiovascular diseases (CVD) recurrence.Methods:This study is a prospective cohort study based on the China Health and Retirement Longitudinal Study (CHARLS) database. The inclusion criteria were participants who underwent the national baseline survey in 2011. The exclusion criteria were as follows: ① female participants; ② participants without BPH/LUTS information at baseline; ③ participants who had not experienced CVD at baseline. The basic characteristics of BPH/LUTS patients and non-BPH/LUTS patients was compared. The Cox proportional hazards model was used to analyze whether BPH/LUTS is a risk factor for CVD recurrence.Results:A total of 903 male patients with stroke or heart disease were finally included, among which there were 324 (36.9%) cases of BPH/LUTS and 579 (64.1%) cases without BPH/LUTS. Patients with BPH/LUTS had a higher proportion of comorbidities such as hypertension [186 cases (57.4%) vs. 165 cases (28.5%), P=0.014], dyslipidemia [100 cases (31.2%) vs. 104 cases (18.2%), P<0.001], diabetes [55 cases (17.2%) vs. 57 cases (9.8%), P=0.002], chronic kidney disease [71 cases (22.2%) vs. 56 cases (9.8%), P<0.001], and chronic lung disease [87 cases (26.9%) vs. 117 cases (20.3%), P=0.029] compared to those without BPH/LUTS. During the 7-year follow-up, there were 341 (37.8%) cases of CVD recurrence, including 319 (35.3%) cases of heart disease recurrence and 38 (4.2%) cases of stroke recurrence. In the multifactorial adjusted Cox regression model, BPH/LUTS was a risk factor for CVD recurrence ( HR=1.41, 95% CI 1.12-1.77, P=0.003) and heart disease recurrence ( HR=1.40, 95% CI 1.10-1.77, P=0.005), while BPH/LUTS was not a risk factor for stroke recurrence ( HR=2.05, 95% CI 0.97-4.32, P=0.058). When stratified by age, in the population aged ≤65 years, BPH/LUTS was a risk factor for CVD ( HR=1.43, 95% CI 1.07-1.91, P=0.002) and heart disease recurrence ( HR=1.40, 95% CI 1.03-1.90, P=0.006), but not for stroke recurrence ( HR=2.16, 95% CI 0.98-6.09, P=0.053). In patients aged >65 years, BPH/LUTS was not a risk factor for CVD ( P=0.110), heart disease ( P=0.096), or stroke recurrence ( P=0.830). Conclusions:BPH/LUTS is closely related to an increased risk of CVD recurrence, especially for CVD patients aged ≤65 years old. Therefore, for elderly male CVD patients aged ≤65 years, preventing and actively treating BPH/LUTS may have significant implications.

Objective:To explore the impact of benign prostatic hyperplasia (BPH)/lower urinary tract symptoms (LUTS) on the risk of cardiovascular diseases (CVD) recurrence.Methods:This study is a prospective cohort study based on the China Health and Retirement Longitudinal Study (CHARLS) database. The inclusion criteria were participants who underwent the national baseline survey in 2011. The exclusion criteria were as follows: ① female participants; ② participants without BPH/LUTS information at baseline; ③ participants who had not experienced CVD at baseline. The basic characteristics of BPH/LUTS patients and non-BPH/LUTS patients was compared. The Cox proportional hazards model was used to analyze whether BPH/LUTS is a risk factor for CVD recurrence.Results:A total of 903 male patients with stroke or heart disease were finally included, among which there were 324 (36.9%) cases of BPH/LUTS and 579 (64.1%) cases without BPH/LUTS. Patients with BPH/LUTS had a higher proportion of comorbidities such as hypertension [186 cases (57.4%) vs. 165 cases (28.5%), P=0.014], dyslipidemia [100 cases (31.2%) vs. 104 cases (18.2%), P<0.001], diabetes [55 cases (17.2%) vs. 57 cases (9.8%), P=0.002], chronic kidney disease [71 cases (22.2%) vs. 56 cases (9.8%), P<0.001], and chronic lung disease [87 cases (26.9%) vs. 117 cases (20.3%), P=0.029] compared to those without BPH/LUTS. During the 7-year follow-up, there were 341 (37.8%) cases of CVD recurrence, including 319 (35.3%) cases of heart disease recurrence and 38 (4.2%) cases of stroke recurrence. In the multifactorial adjusted Cox regression model, BPH/LUTS was a risk factor for CVD recurrence ( HR=1.41, 95% CI 1.12-1.77, P=0.003) and heart disease recurrence ( HR=1.40, 95% CI 1.10-1.77, P=0.005), while BPH/LUTS was not a risk factor for stroke recurrence ( HR=2.05, 95% CI 0.97-4.32, P=0.058). When stratified by age, in the population aged ≤65 years, BPH/LUTS was a risk factor for CVD ( HR=1.43, 95% CI 1.07-1.91, P=0.002) and heart disease recurrence ( HR=1.40, 95% CI 1.03-1.90, P=0.006), but not for stroke recurrence ( HR=2.16, 95% CI 0.98-6.09, P=0.053). In patients aged >65 years, BPH/LUTS was not a risk factor for CVD ( P=0.110), heart disease ( P=0.096), or stroke recurrence ( P=0.830). Conclusions:BPH/LUTS is closely related to an increased risk of CVD recurrence, especially for CVD patients aged ≤65 years old. Therefore, for elderly male CVD patients aged ≤65 years, preventing and actively treating BPH/LUTS may have significant implications.

Objective:To analyze the correlation between the muscle thickness (MT) and echo intensity (EI) of the medial head of gastrocnemius and muscle mass, strength and functional parameters, and to evaluate the diagnostic efficacy of MT and EI for decreased muscle mass.Methods:Fifty-two subjects who visited the Department of Ultrasound Medicine of the Second Affiliated Hospital of Fujian Medical University from March to December 2021 were collected.MT of the medial head of gastrocnemius was measured by two-dimensional ultrasound, and EI values in the region of interest were analyzed using Image J software. Twenty of the subjects were selected to assess the intra-observer and inter-observer agreements in measuring MT and EI values using the intraclass correlation coefficient (ICC) and Altman-Bland analysis was performed. The subjects were divided into healthy control group (30 cases) and decreased muscle mass group (22 cases) according to 2019 Asian Sarcopenia Working Group (AWGS) diagnostic criteria, the differences in MT and EI values between the two groups were compared, and the correlation between MT, EI values and muscle mass, strength, function were analyzed using Pearson correlation analysis. The diagnostic cutoff values of MT and EI values for decreased muscle mass were obtained using ROC curves, and the diagnostic efficacy was compared.Results:①The ICC of MT and EI in medial head of gastrocnemius detected by the same physician and different physicians were greater than 0.97, and the retest reliability was good. ②Compared with the healthy control group, the MT value in medial head of gastrocnemius decreased and the EI value increased in the decreased muscle mass group ( t=4.759, -3.789; all P<0.001). ③The MT value in medial head of gastrocnemius was positively correlated with appendicular skeletal muscle mass (ASM), appendicular skeletal muscle mass index (ASMI), lower limb skeletal muscle mass (LLSM), grip strength, and gait speed ( r=0.579, 0.568, 0.585, 0.495, 0.415; all P<0.001), of which, the correlation with LLSM was the best; ④EI in medial head of gastrocnemius was negatively correlated with ASM and ASMI, LLSM, grip strength, and gait speed, ( r=-0.583, -0.591, -0.590, -0.537, -0.455; all P<0.001), and among which the correlation with ASMI was the best. ⑤The medial head of gastrocnemius MT and EI were 14.33 mm and 58.85 for predicting decreased muscle mass, respectively, and there was no significant difference in the diagnostic performance of the two for decreased muscle mass ( Z=0.734 3, P>0.05). Conclusions:The MT and EI values of the medial head of gastrocnemius can accurately assess muscle mass and can be used in the evaluation of patients with decreased muscle mass.

Artificial Intelligence ; Forecasting ; Humans ; Rare Diseases ; Urinary Bladder Neoplasms/diagnosis*