The Impacts of Maternal Gestational Diabetes Mellitus (GDM) on Fetal Hearts

Objective To evaluate the fetal cardiac function in gestational diabetes mellitus （GDM） pregnancies under different maternal glycemic controls. Methods Forty four GDM mothers received 78 fetal echocardiographic evaluations at three gestational periods （〈28, 28-34 and ≥34 weeks） and were divided into poorly-（DM1） and well-（DM2） controlled groups according to their glycemic control at examination. Seventy uncomplicated mothers were selected as controls. Parameters of fetal cardiac anatomy and function were measured and analyzed. Results GDM fetuses＇ cardiac ventricular walls were thicker than controls＇, and the differences between DM1 and DM2 were not significant except for end-diastolic left ventricular walls. In both GDM groups, the aortic flow velocities increased earlier than pulmonary artery and DM1 fetuses changed earlier than DM2 ones. GDM fetuses＇ left atrial shortening fraction was smaller than the controls＇ in the period of ≥34 weeks and negatively correlated with thicknesses of left ventricular walls and interventricular septum in DM1 fetuses （r=-0.438 and -0.506）. The right ventricular diastolic function in DM1 and DM2 fetuses decreased after the period of 28-34 weeks and in the period of ≥34 weeks respectively. Tel index of both left and right ventricles increased in DM1 group after the period of 〈28 weeks and in DM2 group only in the period of ≥ 34 weeks, with no significant differences between DM1 and DM2 groups in this period. Conclusion Fetuses of GDM mothers showed cardiac function impairments. Good maternal glycemic control may delay the impairments, but cannot reduce the degree. Some cardiac changes in GDM fetuses were similar to those in pregestational diabetic pregnancies except for several parameters and their changing time.

Accuracy of ultrasonography in diagnosis of fetal central nervous system malformation

BACKGROUND Prenatal examination is an important measure for the screening and diagnosis of fetal malformations.AIM To investigate the accuracy of ultrasonography in the diagnosis of fetal central nervous system(CNS)malformations.METHODS One hundred and thirteen pregnant women suspected of having fetal CNS malformations were examined at our hospital from December 2018 to October 2020 using two-dimensional ultrasonography and three-dimensional ultrasonography,respectively.RESULTS According to the pathological results,there were 79 cases of CNS malformations and 34 cases of non-CNS malformations among the 113 pregnant women suspected of having fetal CNS malformation.Fifty-one cases of CNS malformation and 26 cases of non-CNS malformation were detected by two-dimensional ultrasonography,and 73 cases of CNS malformation and 30 cases of non-CNS malformation were detected by three-dimensional ultrasonography.The diagnostic sensitivity(92.41%)and accuracy(91.15%)of three-dimensional ultrasonography were higher than those of two-dimensional ultrasonography(64.56% and 68.14%,respectively)(P=0.000).The specificity of three-dimensional ultrasonography(88.24%)was higher than that of two-dimensional ultrasonography(76.47%);however,the difference was not significant(P=0.203).CONCLUSION Three-dimensional ultrasonography has high application value in the diagnosis of fetal CNS malformations.In addition,the image quality is clear,and the diagnostic sensitivity and accuracy are high.

Ultrasonographic assessment of cardiac function and disease severity in coronary heart disease

BACKGROUND Coronary heart disease(CHD)causes many adverse cardiovascular events and poses a threat to the patient’s health and quality of life.AIM To evaluate ultrasonography for evaluation of cardiac function and lesion degree in patients with CHD.METHODS A total of 106 patients with CHD(study group)and 106 healthy individuals(control group)in our hospital from March 2019 to September 2020 were selected for this study.All subjects were examined by ultrasound,and the mitral orifice’s early-to-late diastolic blood flow velocity ratio(E/A),left ventricular end-diastolic volume(LVDd),and left atrial diameter(LAD)were measured.Values were compared between the study group and healthy group,and the correlation between the ultrasonic parameters of patients with different cardiac function grades and the degree of CHD were assessed.In addition,the ultrasonic parameters of patients with different prognoses were compared after a follow-up for 6 mo.RESULTS E/A(1.46±0.34)of the study group was smaller than that of the control group(1.88±0.44),while LVDd(58.24±5.05 mm)and LAD(43.31±4.38 mm)were larger(48.15±3.93 and 34.94±2.81,respectively;P<0.05).E/A for patients with grade III disease(1.41±0.43)was smaller and their LVDd(60.04±4.21 mm)and LA(44.16±2.79 mm)were larger than those in patients with grade II disease(1.71±0.48,52.18±3.67 mm,and 39.68±2.37,respectively;P<0.05).Patients with grade IV disease had smaller E/A(1.08±0.39)and larger LVDd(66.81±5.39 mm)and LAD(48.81±3.95 mm)than patients with grade II and III disease(P<0.05).In patients with moderate disease,E/A(1.44±0.41)was smaller and LVDd(59.95±4.14 mm)and LAD(45.15±2.97 mm)were larger than in patients with mild disease(1.69±0.50,51.97±3.88 and 38.81±2.56 mm,respectively;P<0.05).In patients with severe disease,E/A(1.13±0.36)was smaller and LVDd(67.70±6.11 mm)and LAD(49.09±4.05 mm)were larger than in patients with moderate disease(P<0.05).E/A was negatively correlated with cardiac function classi-fication and disease severity,while LVDd and LAD were positively correlated with cardiac function classification and disease severity(P<0.05).E/A(1.83±0.51)for patients with good prognosis was higher than that for those with poor prognosis(1.39±0.32),while LVDd(49.60±4.39 mm)and LAD(36.13±3.05 mm)were lower(P<0.05).CONCLUSION The ultrasonic parameters of patients with CHD are abnormal,and differ significantly in patients with different cardiac function grades,lesion degree,and prognosis.

Prenatal assessment of normal fetal pulmonary grey-scale and lung volume by three-dimensional ultrasonography

Objective To quantitatively analyze the fetal lung echo and right lung volume in the third trimester by real-time three-dimensional ultrasound(3-D US)and evaluate the feasibility of fetal lung maturity.Methods A total of 732 women with normal singleton pregnancies between 28 and 42 weeks of gestation underwent ultrasound examination.The 3-D US equipment with a 3.5-5 MHz transabdominal transducer was used for the fetal biometric measurement.The echogenicity ratio between fetal lung and liver was compared.The fetal lung volume was calculated by the rotational multiplanar technique for volume measurement(VOCAL).Results The right fetal lung volume increased with the increase of gestational age with a linear positive correlation(r=0.884,P<0.01).After 34 weeks,the echogenicity ratio of fetal lung to liver was more than 1.1.Conclusion The echogenicity of lung/liver and fetal lung volume could be used as normal fetal predictable indicators for fetal lung maturity.

Valvuloplasty of fetal pulmonary atresia with intact ventricular septum and hypoplastic right heart: Mid-term follow-up results

Objective:This study aimed to analyze and evaluate the results of mid-term follow-up after fetal pulmonary valvuloplasty(FPV)in fetuses with pulmonary atresia with intact ventricular septum(PA/IVS).Methods:From August 31,2018,to May 31,2019,seven fetuses with PA/IVS and hypoplastic right heart were included in this study.All underwent echocardiography by the same specialist and were operated on by the same team.Intervention and echocardiography data were collected,and changes in the associated indices noted during follow-up were analyzed.Results:All seven fetuses successfully underwent FPV.The median gestational age at FPV was 27.54 weeks.The average FPV procedural time was 6 min.Persistent bradycardia requiring treatment occurred in 4/7 procedures.Finally,five pregnancies were successfully delivered,and the other two were aborted.Compared to data before fetal cardiac interventions(FCI),tricuspid valve annulus diameter/mitral valve annulus diameter(TV/MV)and right ventricle diameter/left ventricle diameter(RV/LV)of all fetuses had progressively improved.The maximum tricuspid regurgitation velocity decreased from 4.60 m/s to 3.64 m/s.The average follow-up time was 30.40±2.05 months.During the follow-up period,the diameter of the tricuspid valve ring in five children continued to improve,and the development rate of the tricuspid valve was relatively obvious from 6 months to 1 year after birth.However,the development of the right ventricle after birth was relatively slow.It was discovered that there were individual variations in the development of the right ventricle during follow-up.Conclusion:The findings support the potential for the development of the right ventricle and tricuspid valve in fetuses with PA/IVS who underwent FCI.Development of the right ventricle and tricuspid valve does not occur synchronously during pregnancy.The right ventricle develops rapidly in utero,but the development of tricuspid valve is more apparent after birth than in utero.

Critical evaluation of American categorization of fetal heart rate (FHR) decelerations and three tier classification—Shortcomings, contradictions, remedies and need for debate

Fetal heart rate (FHR) decelerations are the commonest aberrant feature on cardiotocograph (CTG) thus having a major influence on classification ofFHRpatterns into the three tier system. The unexplained paradox of early decelerations (head compression—an invariable phenomenon in labor) being extremely rare [1] should prompt a debate about scientific validity of current categorization. This paper demonstrates that there appear to be major fallacies in the pathophysiological hypothesis (cord compression—baroreceptor mechanism) underpinning of vast majority of (variable?) decelerations. Rapid decelerations during contractions with nadir matching peak of contractions are consistent with “pure” vagal reflex (head compression) rather than result of fetal blood pressure or oxygenation changes from cord compression. Hence, many American authors have reported that the abrupt FHR decelerations attributed to cord compression are actually due to head compression [2-6]. The paper debates if there are major fundamental fallacies in current categorization of FHR decelerations based concomitantly on rate of descent (reflecting putative aetiology?) and time relationship to contractions. Decelerations with consistently early timing (constituting majority) seem to get classed as “variable” because of rapid descent. A distorted unscientific categorization of FHR decelerations could lead to clinically unhelpful three tier classification system. Hence, the current unphysiological classification needs a fresh debate with consideration of alternative models and re-evaluation of clinical studies to test these. Open debate improves patient care and safety. The clue to benign reflex versus hypoxic nature of decelerations seems to be in the timing rather than the rate of descent. Although the likelihood of fetal hypxemia is related to depth and duration ofFHRdecelerations, the cut-offs are likely to be different for early/late/variable decelerations and it seems to be of paramount importance to get this discrimination right for useful visual or computerized system of CTG interpretation.

Association between Fetal Heart Rate Monitoring during Labor and Neonatal Acidosis in Full-Term Newborns: A Retrospective Multicenter Cohort Study

Introduction: Fetal heart rate monitoring during labor is used to study fetal well-being and predict neonatal acidosis of newborn. Fetal heart rate monitoring is analyzed by the obstetrical team and categorized according to the FIGO guidelines. An important limitation of this diagnostic tool is an inter- and intra-observer variability, leading to subjective cardiotocography interpretation and classification. Our objective was to study the association between the categories of fetal heart rate analysis (according to FIGO classification) and neonatal acidosis of full-term newborns. Study design: This is a multicenter retrospective cohort study conducted between 2014 and 2018 in the Grand-Est region. We searched and included retrospectively children hospitalized in a pediatric intensive care unit in one of the participating hospitals with an ICD-10 coding type “P91.6” corresponding to “Hypoxic Ischemic Encephalopathy”. Maternal, pregnancy, delivery, and newborn characteristics were collected and compared by univariate logistic regression with multiple imputation. Odds Ratio and 95% confidence intervals (CI) were calculated using the model and presented. Multiple imputation with m = 100 imputations was tested, using Rubin rules to combine the results. Results: 55 patients were included in the study. Fetal heart rate tracings classified in Category 3 as “pathological” according to FIGO guidelines were significantly associated with an increased risk of severe neonatal acidosis. Late decelerations and bradycardia during labor were associated with severe neonatal acidosis. Conclusion: Severe neonatal acidosis may be suspected by interpretation of fetal heart rate during labor. Fetal bradycardia and late decelerations are predictive of the severity of neonatal acidosis. This study emphasizes the need to screen severe neonatal acidosis and allows the identification of populations most at risk. Repeated team training and upgrading of fetal heart rate study would further reduce the incidence of neonatal acidosis.

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.

Interpretation of British experts’ illustrations of fetal heart rate (FHR) decelerations by Consultant Obstetricians, registrars and midwives: A prospective study—Reasons for major disagreement with experts and implications for clinical practice

Objective: To test the reproducibility of British experts’ (eFM, K2MS, Gibb and Arulkumaran) [1-3] illustrations of fetal heart rate (FHR) decelerations by trained British Obstetricians and midwives. To analyze reasons for any discrepancies by examining factors relating to the participants, British experts’ descriptions and NICE guidelines [4]. Design: Prospective observational study. Setting: National Health Service (NHS) Hospitals. Participants: 38 Obstetric Consultants, 49 registrars and 45 midwives. Methods: Printed questionnaire. Statistical Analysis: Fisher’s Exact test. Results: This largest study of its kind showed almost unbelievably high disconnect between CTG interpretation by experts and participants. 98% - 100% midwives, 80% - 100% Registrars and 74% - 100% Consultants categorized FHR decelerations differently from the five experts’ illustrations/interpretations (p < 0.0001). Remarkably, the three experts’ illustrations of early (supposedly most benign) decelerations were classed as atypical variable by 56% Consultants, 78% Registrars and 99% midwives and the CTGs as pathological by 85% of the participants. Conclusions: The high degree of disagreement with the experts’ illustrations (p < 0.0001) did not appear to be due to participant factors. The immediate reasons seemed to be the conflicting illustrations and heterogeneity of experts’ descriptions. But most importantly, these appeared to stem from non-standardized ambiguous definitions of FHR decelerations and many intrinsic systemic flaws in the current NICE guidelines [4]. The NICE concept of “true uniform” (identical) early and late decelerations seems biologically implausible (a myth) and no examples can be found. Another myth seems to be that early and late decelerations should be gradual. Only very shallow decelerations will look “gradual” on the British CTG. These systemic flaws lead to dysfunctional CTG interpretation increasing intervention as well as impairing diagnosis of fetal hypoxemia. This is because the vast majority of FHR decelerations fall in a single heterogeneous “variable” group with many further classed as “atypical” (pathological) based on disproven and discredited criteria [5-7]. There is increasing evidence in USA that a system with variable decelerations as the majority is clinically unhelpful because of loss of information [5-9]. In the interest of patient care and safety, open debate is necessary regarding a better way forward. Classification of FHR decelerations based primarily and solely on time relationship to contractions appears more scientific and clinically useful.

The Diagnostic Accuracy of Automated Fetal Heart Echocardiography by Five Dimensional Compared to Two-Dimensional Ultrasound in the Second Trimester of Pregnancy

Objective: To compare the diagnostic accuracy of Fetal Intelligent Navigation Echocardiography (FINE) method by Five-Dimensional Ultrasound in automated fetal heart examination to conventional fetal heart examination by Two-Dimensional ultrasound in second trimester pregnancy. Methods: 90 normal singleton pregnancies at the second trimester were enrolled in this prospective study. Fetal heart was examined by Two-Dimensional Ultrasound (2D US) and Five-Dimensional Ultrasound (FINE). Paired numerical data were compared using the paired samples t test. Inter-method agreement for the rating of the quality of cardiac views by 2D US or 5D US was assessed by calculation of the prevalence-adjusted and bias-adjusted kappa coefficient (PABAK). Results: There was no statistical significant difference between fetal heart echocardiography assessment using 2DUS or 5DUS(FINE) (p-value > 0.05). The overall inter-method agreement between 2D US and 5D US was 0.92. The difference in % of satisfactory views between 2D and 5D US was 1.73%, (95% CI - 0.5% to 3.33%) (P-value = 0.144). Conclusion: Our findings indicate the accuracy of 5DUS (FINE) echocardiography in evaluation of normal heart in the second trimester of pregnancy.

An Effective Machine-Learning Based Feature Extraction/Recognition Model for Fetal Heart Defect Detection from 2D Ultrasonic Imageries

Congenital heart defect,accounting for about 30%of congenital defects,is the most common one.Data shows that congenital heart defects have seriously affected the birth rate of healthy newborns.In Fetal andNeonatal Cardiology,medical imaging technology(2D ultrasonic,MRI)has been proved to be helpful to detect congenital defects of the fetal heart and assists sonographers in prenatal diagnosis.It is a highly complex task to recognize 2D fetal heart ultrasonic standard plane(FHUSP)manually.Compared withmanual identification,automatic identification through artificial intelligence can save a lot of time,ensure the efficiency of diagnosis,and improve the accuracy of diagnosis.In this study,a feature extraction method based on texture features(Local Binary Pattern LBP and Histogram of Oriented Gradient HOG)and combined with Bag of Words(BOW)model is carried out,and then feature fusion is performed.Finally,it adopts Support VectorMachine(SVM)to realize automatic recognition and classification of FHUSP.The data includes 788 standard plane data sets and 448 normal and abnormal plane data sets.Compared with some other methods and the single method model,the classification accuracy of our model has been obviously improved,with the highest accuracy reaching 87.35%.Similarly,we also verify the performance of the model in normal and abnormal planes,and the average accuracy in classifying abnormal and normal planes is 84.92%.The experimental results show that thismethod can effectively classify and predict different FHUSP and can provide certain assistance for sonographers to diagnose fetal congenital heart disease.

New Sudanese Reference Chart of Fetal Bi-ometry and Weight Using Ultrasonography

Background: Many centers in Sudan use the reference data for fetal biometry. The recently published population-based reference either overestimated or underestimated the weight of the fetuses. Objective: To establish a national reference for fetal biometry, and weight by gestational age for singleton fetuses in Sudan. Methods: Data were collected on all singleton live births documented in the data collection sheet done at Saudi Hospital from 2015 to 2016 (n = 225). Gestational age estimation was based on the last menstrual period and fetal ultrasound thereafter. Fetal biometry and weight and other 6 fetal weight formulae were assessed. Reference data for fetal growth by gestational age were created. Results: New charts and reference equations are reported in Sudanese population for fetal biparietal diameter, head circumference, abdominal circumference and femur length and fetal weight. Conclusion: We advocate that these reference charts and equations for fetal biometry and weight might be valuable in the clinical use for appropriate ethnic Sudanese.

Creation of high position fetal balloon atrial septoplasty for hypoplastic left heart syndrome and highly restrictive atrial septum: A case report and literature review

Background:Fetal balloon atrial septoplasty(BAS)is performed through the restrictive foramen ovale in fetal cases with established hypoplastic left heart syndrome(HLHS)and an intact or highly restrictive atrial septum(RAS).Methods:In the current report,we present a case of high position BAS in a fetus with HLHS/RAS.Results:Echocardiography confirmed an adequate atrial opening above the foramen ovale and fetal pleural effusion resolved spontaneously 1 day after the procedure.Conclusion:To the best of our knowledge,the creation of a high position hole in the thinner part of the atrial septum,instead of the restrictive tiny hole,has not been reported in fetal cases with HLHS/RAS.

Right heart modified myocardial performance index and ductus venosus spectrum parameters in pre-eclampsia fetuses for predicting adverse pregnancy outcomes

Objective To observe the values of changes of right heart modified myocardial performance index(Mod-MPI)and ductus venosus(DV)spectrum parameters in pre-eclampsia fetuses for predicting adverse pregnancy outcomes.Methods Eighty-one pregnant women diagnosed as pre-eclampsia were prospectively enrolled and divided into severe pre-eclampsia(SPE)group(n=39)and mild pre-eclampsia(MPE)group(n=42),while 85 healthy pregnant women were taken as controls(control group).Fetal right heart function parameters,including right ventricular isovolumetric relaxation time(IRT),isovolumetric contraction time(ICT),ejection time(ET),total spent time(TST),Mod-MPI,tricuspid valve peak flow velocity ratio in early and late diastole(TV-E/A),as well as blood flow velocities in each waveform of DV spectrum(S,V,D,and A wave)were obtained,and the pulsatility index(PI)and the ratio of blood flow velocities in each waveform of the DV(S/V,S/D,S/A,V/D,V/A,D/A)were calculated.Intrauterine fetal distress,preterm delivery,neonatal asphyxia and newborn with low weight were considered as adverse pregnancy outcomes.The correlations of right heart Mod-MPI and TV-E/A with DV parameters in pre-eclampsia fetuses were assessed,and their predictive efficacies for adverse pregnancy outcomes were evaluated for right heart Mod-MPI and DV using the receiver operating characteristics(ROC)and the area under the curves(AUC).Results Compared with control group and MPE group,fetal right heart IRT,ICT and Mod-MPI increased and ET decreased in SPE group(all P<0.05).No significant differences of right heart TST and TV-E/A among 3 groups(both P>0.05).Fetal DV A-wave velocity and V/D values progressively decreased but PI progressively increased in control,MPE and SPE groups(all P<0.05).Fetal right heart Mod-MPI in pre-eclampsia was moderately positively correlated with DV PI(r=0.637,P=0.016),while TV-E/A was weakly negatively correlated with DV V/D(r=-0.355,P=0.043).Adverse pregnancy outcomes were noticed in 59 pre-eclampsia cases.The AUC of fetal right heart Mod-MPI and DV PI for predicting adverse pregnancy outcomes in pre-eclampsia cases was 0.897 and 0.848,respectively,without significant difference(Z=0.460,P=0.400).Conclusion Changes of right heart Mod-MPI and DV spectrum parameters in pre-eclampsia fetuses had high value for predicting adverse pregnancy outcomes.

Unexplained fetal tachycardia:A case report

BACKGROUND This study aimed to explore the possible etiology and treatment of severe fetal tachycardia in the absence of organic disease and provide a reference for clinical management of severe fetal tachycardia.CASE SUMMARY A 29-year-old pregnant woman,with a gravidity 1 parity 0,presented with a fetal heart rate(FHR)of 243 beats per minute during a routine antenatal examination at 31+2 wk of gestation.Before termination of pregnancy at 38 wk of gestation,the FHR repeatedly showed serious abnormalities,lasting more than 30 min.However,the pregnant woman and the fetus had no clinical symptoms,and repeated examination revealed no organic lesions.The mother and the baby were regularly followed up.CONCLUSION This was a case of severe fetal tachycardia with no organic lesions and management based on clinical experience.

The Signal Extraction of Fetal Heart Rate Based on Wavelet Transform and BP Neural Network

This paper briefly introduces the collection and recognition of bio-medical sig nals, designs the method to collect FM signals. A detailed discussion on the sys tem hardware, structure and functions is also given. Under LabWindows/CVI,the ha rdware and the driver do compatible, the hardware equipment work properly active ly. The paper adopts multi threading technology for real-time analysis and make s use of latency time of CPU effectively, expedites program reflect speed, impro ve s the program to perform efficiency. One threading is collecting data; the other threading is analyzing data. Using the method, it is broaden to analyze the sig nal in real-time. Wavelet transform to remove the main interference in the FM a nd by adding time-window to recognize with BP network; Finally the results of c ollecting signals and BP networks are discussed.8 pregnant women’s signals of F M were collected successfully by using the sensor. The correct of BP network rec ognition is about 83.3% by using the above measure.

Point-of-care ultrasonography spotlight:Could venous excess ultrasound serve as a shared language for internists and intensivists?

Point-of-care ultrasonography(POCUS),particularly venous excess ultrasound(VExUS)is emerging as a valuable bedside tool to gain real-time hemodynamic insights.This modality,derived from hepatic vein,portal vein,and intrarenal vessel Doppler patterns,offers a scoring system for dynamic venous congestion assessment.Such an assessment can be crucial in effective management of patients with heart failure exacerbation.It facilitates diagnosis,quantification of congestion,prognostication,and monitoring the efficacy of decongestive therapy.As such,it can effectively help to manage cardiorenal syndromes in various clinical settings.Extended or eVExUS explores additional veins,potentially broadening its applications.While VExUS demonstrates promising outcomes,challenges persist,particularly in cases involving renal and liver parenchymal disease,arrhythmias,and situations of pressure and volume overload overlap.Proficiency in utilizing spectral Doppler is pivotal for clinicians to effectively employ this tool.Hence,the integration of POCUS,especially advanced applications like VExUS,into routine clinical practice necessitates enhanced training across medical specialties.

应用深度学习模型分类正常胎儿心脏超声切面

目的观察4种深度学习(DL)模型分类正常胎儿7类心脏超声切面的价值。方法回顾性纳入200名于孕周18~24+6周接受胎儿心脏超声检查的正常胎儿,按7∶3比例分为训练集(n=140)和测试集(n=60)。收集三血管气管(3VT)切面、心尖四腔心(A4C)切面、主动脉弓长轴切面、双腔静脉切面、左心室流出道(LVOT)切面、三血管(three-vessel,3V)切面及右心室流出道(right ventricular outflow tract,RVOT)切面二维声像图,经过图像预处理后,采用4种DL模型Vision Transformer(ViT)、Data-efficient Image Transformer(DeiT)、Vision-long short term memory(ViL)及Multi-axis Vision Transformer(MaxViT)分别提取图像特征并构建正常胎儿心脏超声切面模型,以准确率、精确率、召回率及F1分数评估各模型于测试集中分类效能,并以梯度加权类激活映射(Grad-CAM)获取激活特征的热力图可视化图像中最具识别特征的区域。结果ViT、DeiT、ViL及MaxViT模型分类测试集中正常胎儿心脏超声切面的效能均表现优秀,其中MaxViT为最优模型,准确率、精确率、召回率及F1分数分别为98.93%、98.93%、98.95%及98.93%。Grad-CAM可视化结果显示,以DL模型分类正常胎儿心脏超声7类切面时,心脏、血管所在区域红色最深,对于分类的贡献最大,模型关注度最高。结论所获4种DL模型分类正常胎儿心脏超声切面均表现优越,尤以MaxViT模型效能最佳,且分类结果的可解释性获得Grad-CAM验证。

右心室-肺动脉耦连在心血管疾病中的应用进展

右心室-肺动脉耦连(RVPAC)是右心室收缩力与后负荷的关系。当右心室功能和肺血管阻力适配时可维持正常RVPAC,而右心室收缩力不足以匹配后负荷时则导致右心室功能障碍。RVPAC在许多心血管疾病病理生理及疾病进展中起重要作用,是患者预后的决定性因素,早期、准确评估RVPAC对于患者病情评估、临床治疗决策、危险分层及预后判断具有重要意义。目前RVPAC的测量方法较多,分为侵入性与非侵入性方法,其中非侵入性方法主要与超声相关。本文就RVPAC的病理机制、测量方法、不同方法优缺点及其在不同心血管疾病中的应用价值进行综述。