ABSTRACT
Objective: To explore the application value of spatiotemporal image correlation (STIC) combined with 5D Heart intelligent navigation echocardiography in ultrasonic screening of normal fetuses. Methods: Ultrasonic screening of fetal heart was performed in 114 normal fetuses using conventional two-dimensional ultrasound and STIC combined with 5D Heart intelligent navigation technique by doctor A. The time taken for image collection using these two methods were recorded and the same nine sections were scored. One week later, image quality of STIC combined with 5D Heart was scored again by doctor A again, also by doctor B. The qualified rate of each section was compared between the two methods, and the consistency of image quality score by the same doctor and different doctors was analyzed. Results: Two fetuses were excluded due to obvious fetal movement and inability to determine the position of the intelligent navigation point. The time of collection of fetal heart volume images and conventional two-dimension ultrasound images were (94.76±35.21)s and (595.88±139.29)s, respectively (P0.05). The qualification rate of the abdomen and left ventricular outflow tract sections using STIC combined with 5D Heart was lower than those using the conventional two-dimension ultrasound (both P0.75). Conclusion: STIC combined with 5D-Heart intelligent navigation technique can shorten the examination time. The quality of the nine fetal heart view images can meet the needs of fetal heart ultrasonic screening, feasible and repeatable in normal fetal heart ultrasonic screening.
ABSTRACT
Objective To investigate the application of fetal intelligent navigation echocardiography ( FINE ,5D Heart) in the diagnosis of conotruncal anomalies ( CTD ) . Methods For prospective study , volume transducer was used to collect the three-dimensional data of fetal heart at 20th - 30th week of gestation . The STIC volume images were processed by three independent physicians of different hierarchies via 5D Heart software afterwards ,and the quality of the STIC diagnosis as well as the consistence among the physicians were analyzed according to the gold standard of postpartum ultrasonic follow-up or surgery results . Results Twelve out of 16 CTD cases were finally included in the study ,and the volume data were analyzed . Senior physicians A could apply 5D Heart diagnosis to achieve 2D-US diagnosis efficiency and gold standard consistently( P = 0 .157 ,0 .083) ,medium and low seniority physician B and C got roughly equal score using 5D Heart diagnosis( P = 0 .705) ,but the diagnostic performance could not reach 2D-US ( P =0 .000 ,0 .000) and gold standard( P = 0 .000 ,0 .000) . Conclusions Fetal cardiac anomalies of CTD can be diagnosed correctly by an expert using 5D Heart ,which has important application prospects in remote consultation , training and scientific research and teaching of congenital heart disease of fetus .
ABSTRACT
Objective To explore the feasibility of intelligent spatiotemporal image correlation-fetal heart navigator(iSTIC-FHN) in the display of the ductal arch view in normal fetuses ,and to compare two-dimensional echocardiography( 2DE) and iSTIC-FHN based measures of ductus arteriosus( DA ) in normal fetuses . Methods Two hundred and eleven normal fetuses with gestational age of 22 - 32 weeks were enrolled . Each fetus had undergone conventional 2DE examination and the iSTIC fetal cardiac volumes were obtained .The volume datasets were analyzed offline using new automatic software ( the Fetal Heart Navigator ,FHN ) . The diameter of DA were measured by 2DE and iSTIC based FHN ,respectively . Pearson's correlation analysis was used to examine the relation between gestational age and diameter of DA for both 2DE and iSTIC-FHN . The measurement of the diameters of ductus arteriosus by two methods were compared . Results The ductal arch view using FHN was successfully obtained in 165(78 .2％ ) cases among 211 normal cases . The comparison of 2DE and iSTIC-FHN in the measurement of the DA was made in 131(62 .1％ ) normal fetuses . Mean 2DE DA increased from ( 3 .00 ± 0 .23) mm at 22 weeks to ( 5 .42 ± 0 .26) mm at 32 weeks . Mean iSTIC-FHN DA increased from ( 2 .96 ± 0 .21) mm at 22 weeks to ( 5 .36 ± 0 .17) mm at 32 weeks . There was a good correlation between the two methods ( Pearson's R 2 = 0 .942 ,P =0 .539) . Bland-Altman plot showed the 95％ limits of agreement was ( - 0 .3287 ,0 .3111) . Conclusions iSTIC-FHN is the potential tool for the evaluation of the ductal arch view .
ABSTRACT
[Objective] To evaluate the value of cardiac screening in the first trimester by using three-dimensional ultrasound with spatiotemporal image correlation (STIC) and to analyze the value of different views for cardiac assessment.[Methods] This study retrospectively reviewed the eighty-seven fetuses with congenital heart diseases which were diagnosed in our institution between January 4th 2013 and June 30th 2016.We compared the detection rate of prenatal ultrasound screening with STIC in the first trimester and the results of postnatal echocardiography or autopsy.In addition,the detection rates of the different planes,including four-chamber view (4CV),outlet of the left ventricle (OLV),outlet of the right ventricle (ORV) and three vessels-trachea view (3VT) were analyzed.[Results] Severe cardiac anomalies were identified in 62 fetuses (71.3％) and non-severe cardiac anomalies were identified in 25 fetuses (28.7％).Seventy-eight cases (89.7％,78/87) with congenital heart diseases could be detected by using three-dimensional ultrasound with STIC.The detection ratesfor 4CV,4CV+OLV,4CV+ORV,4CV+3VT,4CV+OLV+ORV and 4CV+OLV+ORV+3VT were 48.3％,64.4％,66.7％,79.3％,69.0％ and 89.7％ respectively.[Conclusions] Three-dimensional ultrasound with STIC could detect congenital heart disease in the first trinester.The four-chamber view plus outlet of the left ventricle,outlet of the right ventricle and three vessels-trachea view could increase the detection rates of congenital heart diseases.
ABSTRACT
Objective To evaluate the clinical value of fetal intelligent navigation echocardiography (5D Heart) in the display of key diagnostic elements in basic fetal echocardiographic views.Methods Using volume probe,3D volume datasets of the 209 normal singleton fetuses in the second and third trimesters were acquired from a four-chamber view.After processing the datasets by using 5D Heart,9 cardiac diagnostic planes were acquired and then the image qualities of key diagnosis elements of the heart and great vessels were graded by 3 doctors with different experiences of performing fetal echocardiography.Results Two hundred and sixty-two volume datasets were acquired from 209 normal fetuses,including 53 volume datasets collected when cardiac apex was in different directions,satisfactory datasets failed to be acquired in 27 fetuses because of the maternal obesity,the artifacts of fetal movement,or other influencing factors.Two hundred and thirty-one volume datasets of 182 fetuses could be used for 5D Heart analyzing and displaying,and the displaying rate of 9 standard diagnostic views was up to 88.2％.No significant difference was found when comparing the grading results of one observer in different periods of time,the grading results between two experienced doctors and the grading results among the experienced doctor and one relatively inexperienced doctor (P ＞ 0.05).Conclusions 5D Heart has high reliability and repeatability in obtaining fetal echocardiographic diagnosis views and displaying fetal elemental diagnostic information.
ABSTRACT
Objective To investigate whether different initial planes have an effect on examination of the fetal heart by four-dimensional ultrasound with spatiotemporal image correlation (STIC).Methods The standard examination planes of fetal heart were rendered using the heart volume data which were acquired in different initial planes with STIC,then these rendered planes were compared and scored.Results① At the four-chamber initial planes,the reconstruction rate of longitudinal planes of the aortic arch (46.27%) and the pulmonary artery (49.25%) were poor,in addition,the reconstruction showed low quality,and the average scores were 1.366 and 1.403,respectively.② At the left outflow-tract initial planes,reconstruction rates of all planes were above 75%,and there was no statistical difference(P=0.169).Each reconstruction plane had good score,even if the scores of longitudinal planes of the aortic arch/pulmonary artery were low,whose scores reached 1.936 and 2.051,respectively.③ At the initial longitudinal planes of the aortic arch,the reconstruction rate (96.67%) and average reconstruction score (2.617) of longitudinal planes of the pulmonary artery were obviously higher than those of other planes.④ Compared with different initial planes,volume data acquired from left outflow-tract initial planes could lead to high-quality reconstructed planes of all standard examination planes.While four-chamber initial planes acquired high-quality standard examination planes except longitudinal planes of the aortic arch and pulmonary artery.Volume data from initial longitudinal plane of the aortic arch acquired low-quality standard examination planes except longitudinal planes of the aortic arch and longitudinal plane of the pulmonary artery.Conclusions With STIC,sonographers can not only save the volume data for off-line analysis and remote consultation but also choose the better initial planes for themselves during the examination of fetal hearts.
ABSTRACT
Objective To investigate the value of fetal intelligent navigation echocardiography (FINE,5D Heart)in displaying the normal fetal heart structure during 20 to 38 weeks'gestation.Methods Three-dimensional data of fetal heart during 20 to 38 weeks'gestation were acquired using volume transducer prospectively.The STIC volume images were post-processed with 5D heart software by three physicians with different levels of echocardiographic examination and the quality of the section images were analyzed.The intra-and interobserver repeatability was analyzed for the qualification of section images.The correlation between the quality of image and gestational age,body mass index,the placenta position,fetal initial plane position were evaluated using univariate and multivariable logistic regression analysis.Results Two hundreds and five cases of pregnant women were included,and 9 cases were excluded.Volume data acquisition was completed in 180 pregnant women,and the image acquisition rate was 91.8%.The data were assessed twice by three observers and qualified 9 section images were gotten in 156-180 cases(degree of qualification 86.7%-100%).The screening images were satisfied in 152 cases(84.4%),dissatisfied with 28 cases(15.6%).The intra-observer agreement by three observers was absolute concord in four chamber view and abdomen、stomach view;excellent in three vessel trachea view(3VT),five chamber view, left ventricular outflow tract view and superior and inferior vena cava view;good in right ventricular outflow tract view,ductal arch view and aortic arch view.The inter-observer agreement by each of three observers was absolute concord in four chamber view and abdomen、stomach view;excellent in superior and inferior vena cava view;excellent to good in five chamber view,left ventricular outflow tract view,right ventricular outflow tract view;good in 3VT,ductal arch view and aortic arch view.In a univariate analysis,the quality of the image was associated with obesity(χ2=15.338,P =0.000),gestational age(χ2=23.356,P =0.000).Multivariable analysis showed that the obesity(OR=5.587,P =0.001)and gestational age (OR=8.843,P=0.000)were factors affecting the quality of the image.The average time spent in post-processing the images for first time by three observers was(2.77±1.39)min,and there was no significant difference among observers(P > 0.05).Conclusions Fetal intelligent navigation echocardiography (FINE,5D Heart)based on STIC volumes can successfully generate 9 standard fetal echocardiography views in 86.7%-100% of fetuses in the 2nd and 3rd trimesters.This method can simplify examination of the fetal heart and reduce operator dependency.
ABSTRACT
Objective To evaluate the value of fetal intelligent navigation echocardiography ( 5D Heart) and Virtual Intelligent Sonographer Assistance ( VIS-Assistance ? ) in the display of key diagnostic elements in basic fetal echocardiographic views . Methods The heart volumes data of 209 normal fetuses were collected and then the offline analysis was conducted . Processing of the built-in VIS-Assistance ? for each diagnosis plane was optimized . The display success rate and image quality difference of diagnosis images and diagnosis elements before and after the VIS-Assistance ? processing as well as the repeatability between observers were analyzed and compared . Results Totally 262 fetal heart volume data were collected from 209 fetuses ,and 231 volume data of 182 cases could be used for the analysis ,with the acquiring success rate of 88 .2% . For the display rate of 8 diagnosis sections of the 231 fetal heart volume data before and after VIS-Assistance ? processing ,abdominal transverse section at stomach level showed no significant difference ,while the other diagnostic sections showed significant differences ( all P < 0 .05) ,and the display rate after the processing was higher than that before the processing . For the image quality scores of diagnosis sections before and after VIS-Assistance ? processing ,abdominal transverse section at stomach level showed no significant differences ,while other diagnostic sections showed significant differences ( all P < 0 .01) ,and the image quality scores after the processing were higher than those before the processing . According to Bland-Altman analysis ,95% credibility interval of the total scores of 8 sections measured by different observers before and after VIS-Assistance ? were ( - 0 .17/ + 0 .18 ) and ( - 0 .17/ + 0 .18 ) , respectively ,6 .06% (14/231) and 3 .89% (9/231) of the points were outside the 95% credibility interval , and the rest had high consistency . Conclusions 5D Heart and VIS-Assistance ? technology can significantly improve the display rate of diagnostic section and image quality of element structures , and has high reliability and repeatability in obtaining fetal cardiac diagnosis sections and displaying fetal elemental diagnostic information .
ABSTRACT
Objective To evaluate the concordance of two-dimensional echocardiography (2DE) and intelligent spatiotemporal image correlation (iSTIC) in measuring fetal aortic and aortic arch diameters during the second and third trimesters.Methods Data were collected by a prospective cross-sectional study of 140 normal singleton fetuses with the gestational age from 22 to 32 weeks.A total of 6 dimensions of the fetal aortic and aortic arch,including aortic annular diameter (AO),ascending aorta diameter (AAO),aortic arch diameter [AO Arch (INA to LCCA)],aortic arch diameter [AO Arch (LCCA to LSA)],aortic isthmus diameter and descending aorta diameter (DAO),were measured by two different methods.Concordance was assessed by comparing the measurements acquired by iSTIC with those determined by 2DE and depicted by Bland-Altman plots.Inter-and intra-observer variability was evaluated by the intraclass correlation coefficient (ICC) test.Results A total of 137 iSTIC volumes in 140 cases were found to be suitable for further analysis.Good correlation was observed in the measurements that determined by 2D or iSTIC (Pearson's R2 =0.977-0.983).There was no significant difference in the mean values of all the parameters that measured by two methods.Bland-Altman plot showed that the 95％ limits of agreement (LOA) in AO,AAO,AO Arch (INA to LCCA),AO Arch (LCCA to LSA),aortic isthmus diameter and DAO were (-0.1260/+ 0.2299),(-0.1707/+ 0.2241),(-0.1547/+ 0.2190),(-0.1736/+ 0.2024),(-0.1514/+ 0.2039) and (-0.1485/+ 0.2228),respectively.The points in the outside of LOA were 5.11％ (7/137),4.38％ (6/137),5.11％ (7/137),5.84％ (8/137),4.38％ (6/137)and 4.38％ (6/137),respectively.Conclusions iSTIC has a good agreement with 2DE in measuring fetal aortic and aortic arch dimensions during the second and third trimesters.
ABSTRACT
Objective To explore whether the use of spatiotemporal image correlation (STIC) can supply additional information with respect to conventional fetal echocardiography in the prenatal diagnosis of total anomalous pulmonary venous connection(TAPVC).Methods Twenty-five cases diagnosed as TAPVC received fetal echocardiography examination.Four-dimensional volumes from 19 cases of suspected TAPVC were compared to conventional echocardiography.Echocardiographic characteristics were compared with the results of postnatal work-up and pathology.Results TAPVC was found in 25 cases by fetal echocardiography,four cases were isolated TAPVC,21 TAPVC had associated cardiac anomalies.Among them,nine cases were supracardiac types,six cases were infracardiac types,nine cases were intracardiac types and one was indefinite.Three true positivecases of TAPVC were confirmed after birth.Twenty-two choose to terminate pregnancy,9 of which refused to autopsy,and 10 true positive cases of TAPVC were confirmed at autopsy while 3 were considered false-positive cases.4D ultrasound with STIC clearly visualized the anomalous PV confluence and/or the draining vertical vein in all nineteen cases examined.Conclusions STIC may be proposed to identify abnormal venous drainage at the screening level,thus supplying additional information over that provided by 2D fetal echocardiography.
ABSTRACT
Objective To evaluate the application of STIC with segmented analytical method in the diagnosis of fetal atrial septal defect. Method Review the 6600 high-risk pregnant women who accepted PHILIPS iE33 Echocardio graphy examination in our hospital , twelve cases by two-dimensional echocardiographicas atrial septal defect then using real-time STIC method to collect data and offline playback compared with sonography after birth. Results Compared with two-dimensional gray-scale ultrasonography , in the diagnosis of fetal atrial septal defect the coincidence rate of STIC with segmented analytical method was 96.5% and 96.7% respectively , no statistically difference between them (P > 0.03). However, the operating time in STIC with segmented analytical method is shorter, and with the statistically significant (P < 0.03). Conclusion Compared with gray-scale ultrasound , STIC technique combined with segmented analytical method can obtain more valuable information and improve the diagnosis coincidence rate , which is a valuable new technology.
ABSTRACT
Objective To investigate the feasibility of spatiotemporal image correlation(STIC)to assess left ventricular volume in normal fetuses.Methods The volume data acquisition of 162 normal fetuses ranging between 22 and 38 weeks of gestation was achieved by automatic sweep scan.The virtual organ computer-aided analysis(VOCAL Ⅱ)was then used to evaluate end-systolic and end-diastolic volumes of left ventricle and compared with that obtained by M-mode Teichholz formula and 2D biplane Simpson's method.Results Twenty-six fetuses with poor STIC image and 2 twin pregnancies were excluded,the study subjects consisted of 134 singleton fetuses.The highest concordance correlation(r =0.968,for EDV,r =0.956 for ESV;P =0.000)was noted between 2D biplane Simpson' s and STIC VOCAL Ⅱ values with a small bias(0.282 ml for EDV,0.117 ml for ESV).The Bland-Altman analysis showed that STIC VOCAL Ⅱ had the best agreement than M-mode Teichholz formula and 2D biplane Simpson's method.The result showed that there were significant difference between Mmode Teichholz formula and Simpson's method in EDV and ESV(P＜0.05).There were significant differences between M-mode Teichholz formula and STIC VOCAL Ⅱ(P＜0.05).There were no significant differences between STIC VOCAL Ⅱ and Simpson' s method in EDV and ESV.Conclusions There is a good agreement between left ventricular volumes measured either by STIC VOCAL Ⅱ or by 2D biplane Simpson's method.STIC is a novel technique that can be more accurate than conventional Methods in quantification of fetal left ventricular volumes.All the volume data acquisition is achieved by postprocessing facilities.STIC promises to become a new method for left ventricular volume quantification.
ABSTRACT
Objective To establish the reference ranges of the spatial angles among cardiac chambers and great vessels in second and third trimester fetuses measured by spatiotemporal image correlation (STIC).Methods Volume images of 352 normal fetuses from 20 to 38 weeks of gestation were recruited in the study.An off-line analysis of acquired volume datasets was carried out with multiplanar mode.Parameters measured included angles between:(1) the 4-chamber view and the left ventricular long axis view; (2) the left ventricular long axis view and main pulmonary artery; and (3) the ductal arch and aortic arch.The relationships between above-mentioned angles and gestational age were assessed by correlation and regression analysis.Results The angle between the 4-chamber view and the left ventricular long axis view (range:55.7° - 35.7°,mean:45.7° ± 5.12°) was uncorrelated with gestational age (r = 0.03,P = 0.51).In contrast,the angle between the left ventricular long axis view and main pulmonary artery,and the angle between the ductal arch and aortic arch were correlated with gestational age (P ＜ 0.001),and the correlation coefficient was - 0.53 and 0.57 respectively.The best-fit exponential curve regression equations of the angle between the left ventricular long axis view and main pulmonary artery was:Y = 154- 4.24X +0.05X2 ,and the angle between the ductal arch and aortic arch was:Y = - 20.8 + 2.65X - 0.37X2.Conclusions The angles among cardiac chambers and great arteries of fetuses from 20 to 38 weeks of gestation can be quantitatively measured by STIC.The reference ranges provide a reliable quantitative standard to estimate the spatial relationships of the cardiac large arteries of fetuses,which may be clinically useful in prenatal screening congenital heart disease.