To probe into the clinical application value of artificial intelligence in the judgment of the quality standard of mid-sagittal plane ultrasound images for first trimester fetal crown-rump length (CRL).

A total of 1251 midsagittal plane ultrasound images of fetuese at 11-13⁺⁶ weeks of gestation were selected from the database of Shenzhen Maternity & Child Healthcare Hospital from January to December 2022. Using the unified judging results of the image quality standard by the Expert Committee of Prenatal Ultrasound as the golden standard, the performance of an artificial intelligent based quality control model, senior, middle, and junior sonographers in the judgment of the quality standard of mid-sagittal plane ultrasound images for first trimester CRL was assessed by calculating their coincidence rates with the golden standard. The coincidence rates were compared using the Mcnemar-Bowker tests, and weighted Kappa values were applied to analyse the difference and consistency among these results. Time of quality control for each set of images was recorded as one group per 100 images. The Wilcoxon's two samples signed-rank test was applied to compare the difference in the time spent among the four groups.

The coincidence rate of the intelligent quality control model with the golden standard was 90.5%, suggesting a strong consistency to the golden standard (Kappa=0.83, P<0.001), which was slightly lower than that of senior sonographers (90.5% vs 91.1%, χ²=16.40, P<0.001), but superior to that of middle and junior sonographers' (90.5% vs 78.7% vs 68.9%, χ²=100.25, 16.88, P<0.001 for all). The time spent by the intelligent model quality control was significantly less than that by ultrasound physicians [3.57 (3.55, 3.60) s vs 351 (309, 384) s vs 363 (351, 370) s vs 433 (407, 475) s; Z=-3.180, -3.181, and -3.180, respectively, P<0.001 for all].

The intelligent quality control model is accurate and efficient in the judgment of the quality standard of mid-sagittal plane ultrasound images for first trimester CRL.

To establish the reference ranges of fetal prenasal thickness (PT) and nasal bone length (NBL) based on a Chinese population and to assess their predictive value for trisomy 21, 18, and 13 in the second trimester of pregnancy.

Pregnant women who underwent ultrasound during the second trimester were included in this study. The PT and NBL were measured retrospectively in 253 normal fetuses and in 55 fetuses with trisomy 21, 18, and 13.

The mean PT [PT=-4.959+0.565GA-0.01GA² (R²=0.725, P＜0.001)] and NBL [NBL (mm)=-5.656+0.761GA-0.01GA² (R²=0.891, P＜0.001)] both increased with the gestational age, while the PT/NBL ratio (R²=0.004, P=0.376) remained stable. For the fetuses with trisomy 21, 18, and 13, 26.3% (10/38), 36.4% (4/11), and 16.7% (1/6) had a PT below the 5^th percentile, and 42.1% (16/38), 36.4% (4/11), and 33.3% (2/6) had a NBL below the 5^th percentile. The detection rate increased remarkably when PT/NBL ratio were used as a new marker for trisomy 21 fetuses, but not for trisomy 18 and 13 fetuses [65.8% (25/38), 45.5% (5/11), and 16.6% (1/6), respectively].

Fetal PT/NBL ratio could be used as an effective predictive maker for trisomy 21, 18, and 13.

To explore the prenatal ultrasound features of fetal cerebellar cortical dysplasia.

Referring to the relevant literature on cerebellar cortical dysplasia and based on its magnetic resonance imaging findings, anatomy, genetic results, and clinical prognosis, the prenatal ultrasonographic manifestations of 19 fetuses with cerebellar cortical dysplasia diagnosed by ultrasound at Department of Ultrasound, Shenzhen Maternity and Child Healthcare Hospital Affiliated to Nanfang Medical University from January 2017 to December 2020 were summarized.

The prenatal ultrasonogram of cerebellar cortical dysplasia was characterized by abnormal orientation of fissures, and oblique, vertical, or distorted fissures. Genetic tests were performed in 12 cases, of which 9 had positive results and 3 had negative results. Three of the 19 fetuses after birth were followed up to 17 months, 24 months, and 24 months, respectively; the head circumference of one case was larger than that of normal infants of the same age though no obvious abnormality in psychomotor development was found, one case had only fine motor deficiency, and one case had poor outcome, including global developmental delay, mental retardation, and epilepsy. Pregnancy was terminated in all the other 16 cases.

Cerebellar cortical dysplasia can be prenatally diagnosed based on the characteristic ultrasonic features. Cerebellar cortical dysplasia rarely presents as a single finding and is commonly associated with malformations of the brain or other posterior fossa malformations, chromosomal abnormalities, and genetic syndromes. It is difficult to accurately assess the risk of poor prognosis of cerebellar cortical dysplasia.

To determine the variation of cerebellar fissure with gestational weeks in normal fetuses and to establish the numeric reference range for fetal cerebellar fissure at 16-32 weeks.

From June 2019 to December 2020, 644 singleton fetuses with no structural abnormalities diagnosed by prenatal ultrasound were selected as study subjects at Shenzhen Maternal and Child Health Hospital, and divided into two groups: fetuses at 12-15 weeks of gestation and those at 16-32 weeks. The primary fissure and horizontal fissure in two-dimensional images were observed, the number of cerebellar fissures at 16 to 32 weeks of gestation counted, and the morphological changes analyzed. Forty fetuses were randomly sampled and tested for repeatability between two observers. The normal numeric reference range for cerebellar fissure between 16 and 32 weeks of gestation was established. The correlation between the number of cerebellar fissures and gestational age was analyzed by Spearman correlation, and fitting equations were then established.

There was no significant difference in the repeatability test between the two observers. The scatter plots showed that the number of cerebellar fissures were positively correlated with gestational age (r=0.863, 0.698, 0.831, and 0.932, P＜0.001). Quadratic polynomial regression equations were each fitted to the models. The regression equations for the number of cerebellar fissures in three different regions and the total number of fissures were Y=0.974X-0.014X ² -12.65, Y=0.362X-0.005X ² -4.843, Y=-0.125X+0.006X ²+0.285, and Y=1.800X-0.024X ² -22.904, respectively.

Prenatal ultrasound can be used to observe and evaluate the development of fetal cerebellar fissures. The numeric reference range and morphological changes of the normal fetal cerebellar fissures may be useful for prenatal evaluation of the development of the cerebellar cortex.

To explore the sonographic features of fetal hemimegalencephaly (HME) in the second trimester of pregnancy.

The prenatal ultrasonograms of 4 fetuses, which were diagnosed as having HME at Shenzhen Maternal and Child Health Hospital of Southern Medical University from January 2008 to October 2020, were retrospectively reviewed.

The prenatal ultrasonograms of the 4 HME fetuses were characterized as hemisphere asymmetry, shift in midline structure, uneven cortical thickness, and sylvian fissure asymmetry. Among them, 1 case had severe hydrocephalus in the lateral ventricle, 1 had corpus callosum hypoplasia, optic septum hypoplasia, cerebellar hemisphere and vermis hypoplasia, bilateral hydrocephalus, and nasal bone hypoplasia, and 1 had craniocerebral midline cyst, which was consistent with the ultrasound features reported in the literature. All 4 cases were confirmed by MRI.

HME is a rare congenital developmental malformation of the central system with typical prenatal ultrasound findings. Timely prenatal ultrasound screening can provide more information for prenatal counseling and early intervention for pregnant mothers.

To explore sulcus and gyrus alterations on normal fetal brain surface using three-dimensional inverse crystal and realistic vue (3D-ICRV) technique.

This is an retrospective study in which we retrospectively reviewed the 3D volume images of the normal fetal brain (15 to 35 gestational weeks) collected from January 2019 to December 2021 at Shenzhen Maternity and Child Healthcare Hospital Affiliated to Nanfang Medical University. 3D volume images were acquired in the transthalamic or transventricular plane on transabdominal ultrasonography and post-processed by 3D-ICRV. Volume images that were able to demonstrate the primary sulci and gyri on fetal brain surface clearly were selected. The morphology of the sulcus and gyrus and the alteration of apical angle on the Sylvian fissure were recorded according to the order of gestational weeks. Postnatal follow-up was performed in all cases.

During the study period, 300 ultrasound examinations were reviewed, and 588 3D volume images from 294 fetuses (n=294; median gestational weeks, 26.93; median maternal age, 31.22 years) were qualified. The Sylvian fissure was the first structure recognized between 15 to 25 gestational weeks. From 26 to 35 gestational weeks, the primary sulci and gyri emerged gradually. At 26 weeks, the central sulcus and superior temporal sulcus appear first, and then the superior frontal sulcus, intraparietal sulcus, inferior frontal sulcus and inferior temporal sulcus appear successively. Gestational weeks of 25 to 30 was the best time window to visualize the primary sulcus and gyrus on brain surface. Follow-up showed no abnormality.

Fetal brain surface configuration can be visualized through 3D-ICRV. Understanding these changes is of great significance in evaluating the maturity of cerebral cortex development as well as prenatal consultation and further examination for clinical management.

To explore the prenatal ultrasound imaging features and clinical prognosis of fetal solitary cerebellar hemorrhage (CBH).

The prenatal ultrasonographic manifestations of seven fetuses diagnosed with CHB detected by ultrasonography at Shenzhen Maternity and Child Healthcare Hospital Affiliated to Southern Medical University from January 2012 to August 2021 were reviewed and compared with the MRI image features. The prenatal diagnosis and clinical prognosis of CHB were then analyzed based on the relevant literature.

Abnormal echoes were found in the cerebellum of seven fetuses by prenatal ultrasound, of which five (71.4%) showed hyperecho in the cerebellum, two (28.6%) showed uneven mixed hypoecho and anecho, and all had bilateral cerebellar hemisphere asymmetry. Color Doppler showed no signal of blood flow in tissues with abnormal echo. Antenatal MRI examination of two cases showed right CHB and right cerebellar hypoplasia. The early ultrasonic manifestations of fetal CHB were hyperecho in the cerebellum and silent shadow behind. After 1-2 weeks, the hematoma was gradually absorbed and partially liquefied, and hyperecho was reduced to cystic anecho or uneven hypoecho and anecho. After 1 month, the hematoma was absorbed and the hyperecho or mixed echo disappeared, with asymmetrical bilateral cerebellar hemispheres and hypoplastic cerebellar hemispheres. In terms of pregnancy outcome and follow-up, there were three (42.9%) cases of induced labor, two cases of fetal death (28.6%) in utero, and two cases of normal neurodevelopment in seven months of follow-up postpartum.

Ultrasound combined with MRI can improve the prenatal diagnostic accuracy for CHB. The long-term neurological follow-up results of children with a prenatal diagnosis of CBH can provide data support in parental counseling and improve the pregnancy outcome of fetuses with prenatal CHB.

To evaluate the diagnostic value of the normal reference range obtained by two-dimensional axial planes in abnormal fetal corpus callosum.

Between June 2018 and July 2020, 670 normal and 77 abnormal fetuses at a gestational age of 20 weeks to full term were evaluated by two-dimensional axial planes at Shenzhen Maternity & Child Healthcare Hospital, Southern Medical University. The diagnosis of abnormal fetuses was confirmed by pre-natal MRI/post-natal ultrasound/ autopsy examinations. The abnormal fetuses were divided into four groups: complete agenesis of the corpus callosum, partial agenesis of the corpus callosum, hypoplasia of the corpus callosum, and hypoplasia of the corpus callosum. The largest length of the corpus callosum, the anteroposterior and mediolateral diameters of the genu, body, and splenium, and anterior and posterior angles of the genu and splenium were measured. Non-parametric test was used to compare the parameters between the normal and abnormal groups, and compound scatter plots of all parameters were drawn.

There was no significant difference in maternal age [30 (28, 33) years vs 30 (26, 34) years] and gestational weeks [28 (24, 32) weeks vs 25 (23, 28) weeks] between the 670 normal pregnant women and the 77 abnormal pregnant women (P＞0.05). In the normal group, the success rate in obtaining satisfactory axial planes reached 100%, but it was only 13.9% in obtaining satisfactory sagittal planes in the normal group; the corresponding figures in the abnormal cases were 100% and 62.3%. The anterior and posterior angles of the genu, anterior and posterior angles of the splenium, anteroposterior diameter of the genu, the largest length of the corpus callosum, mediolateral diameter of the genu, anteroposterior diameter of the splenium, anteroposterior diameter of the splenium, mediolateral diameter of the body, and anteroposterior diameter of the body in the abnormal group were significantly lower than those in the normal group [48.00° (0°, 80.68°) vs (66.76°±10.79°); 44.00° (0°, 79.30°) vs 65.10° (49.30°, 83.00°); 41.90° (0°, 69.37°) vs (64.31°±10.27°); 38.50° (0°, 72.10°) vs (65.38°±11.65°); 0.21 (0, 0.40) cm vs (0.44±0.09) cm; 1.10 (0, 2.78) cm vs 3.33 (2.11, 4.20) cm; 1.18 (0, 2.23) cm vs 1.81 (1.22, 2.46) cm; 1.21 (0, 2.20) cm vs (2.02±0.40) cm; 0.18 (0, 0.35) cm vs 0.45 (0.32, 0.63) cm; 0.42 (0, 1.01) cm vs 0.62 (0.40, 0.98) cm; and 0.75 (0, 1.87) cm vs (1.86±0.33) cm, respectively] (Z=-8.959, -8.650, -9.839, -9.993, -12.812, -13.668, -7.343, -10.521, -12.145, -5.260, and -14.034, respectively, P＜0.001). The compound scatter plots showed that the largest length of the corpus callosum was significantly shorter, and the thickness of the genu and splenium with HpCC (thin) was significantly lower in abnormal fetuses than in normal fetuses, showing the parameter measurements below the 5th percentile line of normal values. There was no significant difference in other parameters on the compound scatter plots.

Prenatal two-dimensional axial planes can well evaluate the corpus callosum structure, which is valuable for the evaluation of fetuses with abnormal corpus callosum. In particular, the largest length of the corpus callosum is suitable for the evaluation of most abnormal corpus callosum cases. The anteroposterior diameters of the genu and splenium are useful for the diagnosis of HpCC (thin). Other parameters are of little significance for the diagnosis of corpus callosum abnormalities.

To evaluate the clinical and ultrasonic features of fetal sacrococcygeal teratoma (SCT) of different histological types.

A retrospective analysis was performed on a total of 34 patients with SCT that was diagnosed from January 2015 to December 2016 at Beijing Obstetrics and Gynecology Hospital, Capital Medical University. These cases were divided into either a mature group or an immature group based on the histological type. Data (maternal age, gestational weeks, and tumorous sonographic features) were analyzed by the Fisher exact test, t-test, Chi-square, and Mann-Whitney U test.

No statistical differences were observed in terms of maternal age, gestational week, fetal sex, and Altman's classification between the mature (n=23) and immature groups (n=11). Among all the ultrasonic features, mass volume of the immature group was significantly bigger than that of the mature group [16.19 (8.47, 34.03) cm³ vs 186.20 (90.41, 445.35) cm³, Z=-3.332, P=0.001]. The main manifestation of the mature group was cystic echo (56.5%, 13/23), while that of the immature group was cystic-solid echo (100%, 11/11); the difference between the two groups was statistically significant (P=0.002). Color Doppler ultrasound showed that three cases in the mature group and ten cases in the immature group were detected with blood flow signal; the difference in the number of cases with blood flow signal was statistically significant between the two groups (P＜0.001). Scanning of the tissues surrounding the mass and other fetal systems showed that five cases in the mature group and seven cases in the immature group suffered from complications; there was a statistically significant difference between the two groups (P=0.026).

The clinical features and prenatal ultrasonic images of fetal SCT vary with different pathological types. Cystic-solid echogenic masses, bigger mass volume, rich blood flow signals, and suffering from complications are important signs for the diagnosis of immature fetal SCT. Early diagnosis by using prenatal ultrasound can provide a good basis for perinatal management.

To assess the clinical value of online artificial intelligent quality control system in assessing obstetric ultrasound images.

A total of 374 191 images of 15 640 obstetric ultrasound cases from 998 doctors in 137 hospitals in Shenzhen and Chongqing were selected from January 1 to June 30, 2021, and the quality of the images was evaluated using online artificial intelligent quality control system. Based on each image quality assessment result, the proportion of different standard levels of all planes was calculated. The appeal results were also recorded to observe the accuracy of the system. To survey the efficiency of the system, paired sample t test was used to compare the time spent by intelligent quality control and manual quality control.

The overall standard rate, substandard rate, and nonstandard rate were 81.18%, 12.06%, and 6.76%, respectively. A total of 285 appealed images (0.076%) were reviewed by authoritative experts, who confirmed the initial diagnosis in 126 images (44.21%), and did not support the initial diagnosis in 159 images (55.79%); the accuracy of the system reached 99.96% (374 032/374 191). The average time spent by intelligent quality control for 100 images was (32.7±5.1) s, significantly shorter than that spent by manual quality control by two ultrasound physicians [(705.3±37.2) s and (724.6±40.4) s, t=62.667 and 56.396, respectively, P＜0.001].

The intelligent quality control system of obstetric ultrasound images allows the quality control to be performed objectively, accurately, and efficiently, which is of great significance to guide the improvement of image quality.

To evaluate the clinical value of prenatal three-dimensional (3D) inversion imaging of the sylvian fissure, parietal occipital sulcus, and calcarine sulcus.

The sylvian fissure, parietal occipital sulcus, and calcarine sulcus were assessed prospectively by 3D ultrasound (3D-US) inversion imaging and two-dimensional ultrasound (2D-US) in 133 fetuses with no structural abnormalities between 19 and 36⁺⁶ weeks of gestation at Shenzhen Maternity and Child Healthcare Hospital from June 2019 to July 2020. The examinations were divided into six groups according to gestational age. For each group, these sulci were assessed as absent or present. The changes of these 3D sulci imaging were observed in different groups. Thirty fetuses were randomly selected for intra- and inter-observer reproducibility testing by pearson chi-square test.

Along with the increase of the gestation age in different groups, the rates of 2D and 3D examinations to display the structures both became higher and higher. 2D-US and 3D-US showed 100% of sylvian fissure, parietal occipital sulcus, and calcarine sulcus at 22, 23, and 24~36 weeks, respectively. The display rate had no statistical difference between 2D-US and 3D inverted ultrasound in the six groups (P＞0.05). However, 3D inversion ultrasound was more vivid and intuitive than 2D-US. No significant intra- and inter-observer variability was found.

3D inversion imaging can be used to better observe the morphological and spatial changes of the sylvian fissure, parietal occipital sulcus, and calcarine sulcus during middle and late pregnancy, which is an intuitive supplementary means for prenatal evaluation of cerebral cortex development and provides a new idea for future teaching and research of cerebral cortex development.

To propose a novel prenatal sonographic diagnostic method for fetuses with subarachnoid hemorrhage (SAH).

This was a retrospective study of five fetuses with SAH diagnosed by ultrasound and MRI at Shenzhen Maternity and Child Healthcare Hospital between 2018 and 2021, in which we systematically analyzed and summarized the prenatal ultrasound diagnostic strategy and sonographic features in fetal SAH.

The typical ultrasound manifestations of SAH were: (1) uniform hyperechoic area in the sylvian fissure (n=5); (2) enhanced echo in the surface sulci including the subfrontal sulcus, superior temporal sulcus, and parieto-occipital sulcus (n=4); (3) evenly strong echo in the supracarpal cistern, annular cistern, and interstellar cistern (n=3); (4) hyperecho in anterior and posterior longitudinal fissures, sagittal sinus, or sinus confluence (n=2); (5) subarachnoid thickening with strong echo through the sagittal plane of the spina conus (n=1); (6) hemorrhage at other parts of the brain, causing cerebral parenchymal or intraventricular hemorrhage (n=4); (7) increased blood flow velocity in the middle cerebral artery (n=4); and (8) no blood flow signal in the hemorrhage focus on CDFI.

Typical uniform hyperechoic manifestations of the sylvian fissure are an important clue for prenatal ultrasound to detect and diagnose fetal SAH. It is suggested to identify uniform hyperechoic manifestations on the brain surface including the sulci, cistern, and other parts of the brain. Such manifestations are reliable diagnostic signs to identify whether SAH is combined with other intracranial hemorrhage.