Prenatal diagnosis of fetuses with absent/hypoplastic nasal bone in second-trimester using chromosomal microarray analysis.

BACKGROUND: Pathogenic copy number variants (pCNVs) are associated with fetal ultrasound anomalies, which can be efficiently identified through chromosomal microarray analysis (CMA). The primary objective of the present study was to enhance understanding of the genotype-phenotype correlation in fetuses exhibiting absent or hypoplastic nasal bones using CMA. METHODS: Enrolled in the present study were 94 cases of fetuses with absent/hypoplastic nasal bone, which were divided into an isolated absent/hypoplastic nasal bone group (n = 49) and a non-isolated group (n = 45). All pregnant women enrolled in the study underwent karyotype analysis and CMA to assess chromosomal abnormalities in the fetuses. RESULTS: Karyotype analysis and CMA detection were successfully performed in all cases. The results of karyotype and CMA indicate the presence of 11 cases of chromosome aneuploidy, with trisomy 21 being the most prevalent among them. A small supernumerary marker chromosome (sSMC) detected by karyotype analysis was further interpreted as a pCNV by CMA. Additionally, CMA detection elicited three cases of pCNVs, despite normal findings in their karyotype analysis results. Among them, one case of Roche translocation was identified to be a UPD in chromosome 15 with a low proportion of trisomy 15. Further, a significant difference in the detection rate of pCNVs was observed between non-isolated and isolated absent/hypoplastic nasal bone (24.44% vs. 8.16%, p < .05). CONCLUSION: The present study enhances the utility of CMA in diagnosing the etiology of absent or hypoplastic nasal bone in fetuses. Further, isolated cases of absent or hypoplastic nasal bone strongly suggest the presence of chromosomal abnormalities, necessitating genetic evaluation through CMA.

Automatic segmentation of 15 critical anatomical labels and measurements of cardiac axis and cardiothoracic ratio in fetal four chambers using nnU-NetV2.

BACKGROUND: Accurate segmentation of critical anatomical structures in fetal four-chamber view images is essential for the early detection of congenital heart defects. Current prenatal screening methods rely on manual measurements, which are time-consuming and prone to inter-observer variability. This study develops an AI-based model using the state-of-the-art nnU-NetV2 architecture for automatic segmentation and measurement of key anatomical structures in fetal four-chamber view images. METHODS: A dataset, consisting of 1,083 high-quality fetal four-chamber view images, was annotated with 15 critical anatomical labels and divided into training/validation (867 images) and test (216 images) sets. An AI-based model using the nnU-NetV2 architecture was trained on the annotated images and evaluated using the mean Dice coefficient (mDice) and mean intersection over union (mIoU) metrics. The model's performance in automatically computing the cardiac axis (CAx) and cardiothoracic ratio (CTR) was compared with measurements from sonographers with varying levels of experience. RESULTS: The AI-based model achieved a mDice coefficient of 87.11% and an mIoU of 77.68% for the segmentation of critical anatomical structures. The model's automated CAx and CTR measurements showed strong agreement with those of experienced sonographers, with respective intraclass correlation coefficients (ICCs) of 0.83 and 0.81. Bland-Altman analysis further confirmed the high agreement between the model and experienced sonographers. CONCLUSION: We developed an AI-based model using the nnU-NetV2 architecture for accurate segmentation and automated measurement of critical anatomical structures in fetal four-chamber view images. Our model demonstrated high segmentation accuracy and strong agreement with experienced sonographers in computing clinically relevant parameters. This approach has the potential to improve the efficiency and reliability of prenatal cardiac screening, ultimately contributing to the early detection of congenital heart defects.

Prenatal diagnosis of interrupted aortic arch using high-definition flow render mode and spatiotemporal image correlation.

OBJECTIVES: To evaluate the clinical utility of two dimensional (2D) ultrasound combined with spatiotemporal image correlation (STIC) in diagnosing interrupted aortic arch (IAA) in fetal life. METHODS: A total of 53 cases of fetal IAA were diagnosed using 2D ultrasound combined with STIC, and 53 normal fetuses of the same gestational week were selected. These cases were retrospectively analyzed to assess the utility of employing 2D ultrasound combined with STIC in the diagnosis of IAA. RESULTS: 2D ultrasound combined with STIC detected 22 cases of type A IAA, 24 cases of type B IAA, and seven cases of type C IAA. Furthermore, combining 2D ultrasound with STIC enabled dynamic visualization of the IAA, aiding in prenatal diagnosis. The diagnostic coincidence rate of IAA was found to be higher in the HD-flow combined with STIC than that in the 2D combined with HD-flow. CONCLUSION: HD-flow combined with STIC can assist in diagnosing fetal IAA, and this technique has important clinical value.

The value of fetal magnetic resonance imaging in diagnosis of congenital anomalies of the fetal body: a systematic review and meta-analysis.

OBJECTIVES: To undertake a systematic review to assess the accuracy of fetal MRI in diagnosis of non-CNS congenital anomalies of the fetal body in comparison with antenatal ultrasound when correlated to postnatal diagnosis. METHODS: Searches were conducted from electronic databases, key journals and reference lists for eligible papers. Inclusion criteria was original research studies comparing the diagnostic results of antenatal ultrasound, fetal MRI and final postnatal diagnosis via imaging, surgery or post-mortem testing. Studies of CNS anomalies were excluded. Studies were assessed for risk of bias by two reviewers working independently and data was then extracted by a single reviewer. RESULTS: 12 studies were included with a total of 361 eligible patients who underwent USS and MRI and had a postnatal diagnosis. USS alone had a diagnostic accuracy of 60.6% whereas MRI had an improved diagnostic accuracy of 86.4%. The overall odds ratio was 0.86 (CI 0.202-1.519 and p-value < 0.01). CONCLUSION: Fetal MRI makes a significant contribution to accurate diagnosis of congenital abnormalities of the fetal body; especially in genito-urinary anomalies. More research is needed to improve the evidence base for the role of fetal MRI in diagnosis of congenital anomalies in other body systems.

RTSeg-net: A lightweight network for real-time segmentation of fetal head and pubic symphysis from intrapartum ultrasound images.

The segmentation of the fetal head (FH) and pubic symphysis (PS) from intrapartum ultrasound images plays a pivotal role in monitoring labor progression and informing crucial clinical decisions. Achieving real-time segmentation with high accuracy on systems with limited hardware capabilities presents significant challenges. To address these challenges, we propose the real-time segmentation network (RTSeg-Net), a groundbreaking lightweight deep learning model that incorporates innovative distribution shifting convolutional blocks, tokenized multilayer perceptron blocks, and efficient feature fusion blocks. Designed for optimal computational efficiency, RTSeg-Net minimizes resource demand while significantly enhancing segmentation performance. Our comprehensive evaluation on two distinct intrapartum ultrasound image datasets reveals that RTSeg-Net achieves segmentation accuracy on par with more complex state-of-the-art networks, utilizing merely 1.86 M parameters-just 6 % of their hyperparameters-and operating seven times faster, achieving a remarkable rate of 31.13 frames per second on a Jetson Nano, a device known for its limited computing capacity. These achievements underscore RTSeg-Net's potential to provide accurate, real-time segmentation on low-power devices, broadening the scope for its application across various stages of labor. By facilitating real-time, accurate ultrasound image analysis on portable, low-cost devices, RTSeg-Net promises to revolutionize intrapartum monitoring, making sophisticated diagnostic tools accessible to a wider range of healthcare settings.

Malformations of cortical development: Fetal imaging and genetics.

BACKGROUND: Malformations of cortical development (MCD) are a group of congenital disorders characterized by structural abnormalities in the brain cortex. The clinical manifestations include refractory epilepsy, mental retardation, and cognitive impairment. Genetic factors play a key role in the etiology of MCD. Currently, there is no curative treatment for MCD. Phenotypes such as epilepsy and cerebral palsy cannot be observed in the fetus. Therefore, the diagnosis of MCD is typically based on fetal brain magnetic resonance imaging (MRI), ultrasound, or genetic testing. The recent advances in neuroimaging have enabled the in-utero diagnosis of MCD using fetal ultrasound or MRI. METHODS: The present study retrospectively reviewed 32 cases of fetal MCD diagnosed by ultrasound or MRI. Then, the chromosome karyotype analysis, single nucleotide polymorphism array or copy number variation sequencing, and whole-exome sequencing (WES) findings were presented. RESULTS: Pathogenic copy number variants (CNVs) or single-nucleotide variants (SNVs) were detected in 22 fetuses (three pathogenic CNVs [9.4%, 3/32] and 19 SNVs [59.4%, 19/32]), corresponding to a total detection rate of 68.8% (22/32). CONCLUSION: The results suggest that genetic testing, especially WES, should be performed for fetal MCD, in order to evaluate the outcomes and prognosis, and predict the risk of recurrence in future pregnancies.

Comparing three cardiothoracic ratio measurement techniques and creating multivariable scoring system to predict Bart's hydrops fetalis at 17-22 weeks' gestation.

To assess the diagnostic performance of three cardiothoracic (CT) ratio techniques, including diameter, circumference, and area, for predicting hemoglobin (Hb) Bart's disease between 17 and 22 weeks' gestation, and to create a multivariable scoring system using multiple ultrasound markers. Before invasive testing, three CT ratio techniques and other ultrasound markers were obtained in 151 singleton pregnancies at risk of Hb Bart's disease. CT diameter ratio demonstrated the highest sensitivity among the other techniques. Significant predictors included CT diameter ratio > 0.5, middle cerebral artery-peak systolic velocity (MCA-PSV) > 1.5 multiples of the median, and placental thickness > 3 cm. MCA-PSV exhibited the highest sensitivity (97.8%) in predicting affected fetuses. A multivariable scoring achieved excellent sensitivity (100%) and specificity (84.9%) for disease prediction. CT diameter ratio exhibited slightly outperforming the other techniques. Increased MCA-PSV was the most valuable ultrasound marker. Multivariable scoring surpassed single-parameter analysis in predictive capabilities.

Deep learning prediction of renal anomalies for prenatal ultrasound diagnosis.

Deep learning algorithms have demonstrated remarkable potential in clinical diagnostics, particularly in the field of medical imaging. In this study, we investigated the application of deep learning models in early detection of fetal kidney anomalies. To provide an enhanced interpretation of those models' predictions, we proposed an adapted two-class representation and developed a multi-class model interpretation approach for problems with more than two labels and variable hierarchical grouping of labels. Additionally, we employed the explainable AI (XAI) visualization tools Grad-CAM and HiResCAM, to gain insights into model predictions and identify reasons for misclassifications. The study dataset consisted of 969 ultrasound images from unique patients; 646 control images and 323 cases of kidney anomalies, including 259 cases of unilateral urinary tract dilation and 64 cases of unilateral multicystic dysplastic kidney. The best performing model achieved a cross-validated area under the ROC curve of 91.28% ± 0.52%, with an overall accuracy of 84.03% ± 0.76%, sensitivity of 77.39% ± 1.99%, and specificity of 87.35% ± 1.28%. Our findings emphasize the potential of deep learning models in predicting kidney anomalies from limited prenatal ultrasound imagery. The proposed adaptations in model representation and interpretation represent a novel solution to multi-class prediction problems.

Importance of Prenatal Diagnosis of Ileal Atresia in Gestational Diabetes Cases.

BACKGROUND Maldevelopment of the fetal bowel can result in the rare condition of intestinal atresia, which results in congenital bowel obstruction. This report describes a case of prenatal diagnosis of fetal ileal atresia at 22 weeks' gestation. CASE REPORT Here, we present a 24-year old woman who was 22 weeks into her first pregnancy when she underwent routine fetal ultrasound. She was diagnosed with gestational diabetes mellitus. Her body mass index was normal and she had normal weight gain. The ultrasonographic examination performed revealed a hyperechoic bowel and a small dilatation of the bowel. The couple was counselled for possible intestinal atresia and its postnatal implications. At 33 weeks of gestation, polyhydramnios appeared, and the intestinal distension was much more pronounced, with hyperechoic debris in the intestinal lumen (succus-entericus). After birth, surgery was performed and we concluded the patient had type II atresia, which was surgically treated. CONCLUSIONS This report has highlighted the importance of antenatal ultrasound in detecting fetal abnormalities, and has shown that rare conditions such as intestinal atresia can be accurately diagnosed and successfully managed. Surgical correction, if implemented promptly after stabilizing the general condition, can have a relatively good prognosis. Coexisting fetal ileal atresia and gestational diabetes mellitus are rare occurrences, which can make each condition even more difficult to treat.

[Congenital diaphragmatic hernia : Imaging-from diagnosis to aftercare]. / Kongenitale Zwerchfellhernie : Bildgebung von der Diagnose bis zur Nachsorge.

STANDARD RADIOLOGICAL METHODS: Fetal: Ultrasound and magnetic resonance imaging (MRI); postnatal: conventional X­ray diagnostics, computed tomography (CT) and MRI. METHODICAL INNOVATIONS: MRI-based lung ventilation and perfusion measurement. PRACTICAL RECOMMENDATIONS: Lifelong follow-up care should be provided, in which radiology is part of the treatment team.

[Developmental disorders of the gastrointestinal tract]. / Entwicklungsstörungen des Gastrointestinaltrakts.

BACKGROUND: Developmental disorders of the gastrointestinal tract comprise a broad spectrum of congenital malformations of different etiologies and locations from the mouth to the anus. METHODS: The authors present the most important malformations of the gastrointestinal tract on the basis of basic and current reviews. RESULTS: Gastrointestinal developmental disorders occur both sporadically and in connection with malformation syndromes. Symptoms are highly variable and range from postnatal emergencies to asymptomatic abnormalities, which may be incidental radiological findings. Prenatal ultrasound examinations can often identify gastrointestinal developmental disorders at an early stage. Here, fetal magnetic resonance imaging can be a useful addition to the diagnostic process. In the first few days of life, simple X­ray overview images, supplemented by images after the administration of contrast medium, are often sufficient. CONCLUSION: Many patients with a malformation of the gastrointestinal tract require lifelong medical care, so that not only pediatric radiologists need specific knowledge about this group of diseases.

Routine Antenatal Echocardiography in High-Prevalence Areas of Rheumatic Heart Disease: A WHO-Guideline Systematic Review.

Background: Rheumatic Heart Disease (RHD) is the most common cause of valvular heart disease worldwide. Undiagnosed or untreated RHD can complicate pregnancy and lead to poor maternal and fetal outcomes and is a significant factor in non-obstetric morbidity. Echocardiography has an emerging role in screening for RHD. We aimed to critically analyse the evidence on the use of echocardiography for screening pregnant women for RHD in high-prevalence areas. Methods: We searched MEDLINE and Embase to identify the relevant reports. Two independent reviewers assessed the reports against the eligibility criteria in a double-blind process. Results: The searches (date: 4 April 2023) identified 432 records for screening. Ten non-controlled observational studies were identified, five using portable or handheld echocardiography, comprising data from 23,166 women. Prevalence of RHD varied across the studies, ranging from 0.4 to 6.6% (I2, heterogeneity >90%). Other cardiac abnormalities (e.g., congenital heart disease and left ventricular systolic dysfunction) were also detected <1% to 2% of cases. Certainty of evidence was very low. Conclusion: Echocardiography as part of antenatal care in high-prevalence areas may detect RHD or other cardiac abnormalities in asymptomatic pregnant women, potentially reducing the rates of disease progression and adverse labor-associated outcomes. However, this evidence is affected by the low certainty of evidence, and lack of studies comparing echocardiography versus standard antenatal care. Prospective Registration: PROSPERO 2022 July 4; CRD42022344081 Available from: https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=344081. Research question: 'In areas with a high prevalence of rheumatic heart disease, should handheld echocardiography be added to routine antenatal care?'

Deep-learning computer vision can identify increased nuchal translucency in the first trimester of pregnancy.

OBJECTIVE: Many fetal anomalies can already be diagnosed by ultrasound in the first trimester of pregnancy. Unfortunately, in clinical practice, detection rates for anomalies in early pregnancy remain low. Our aim was to use an automated image segmentation algorithm to detect one of the most common fetal anomalies: a thickened nuchal translucency (NT), which is a marker for genetic and structural anomalies. METHODS: Standardized mid-sagittal ultrasound images of the fetal head and chest were collected for 560 fetuses between 11 and 13 weeks and 6 days of gestation, 88 (15.7%) of whom had an NT thicker than 3.5 mm. Image quality was graded as high or low by two fetal medicine experts. Images were divided into a training-set (n = 451, 55 thick NT) and a test-set (n = 109, 33 thick NT). We then trained a U-Net convolutional neural network to segment the fetus and the NT region and computed the NT:fetus ratio of these regions. The ability of this ratio to separate thick (anomalous) NT regions from healthy, typical NT regions was first evaluated in ground-truth segmentation to validate the metric and then with predicted segmentation to validate our algorithm, both using the area under the receiver operator curve (AUROC). RESULTS: The ground-truth NT:fetus ratio detected thick NTs with 0.97 AUROC in both the training and test sets. The fetus and NT regions were detected with a Dice score of 0.94 in the test set. The NT:fetus ratio based on model segmentation detected thick NTs with an AUROC of 0.96 relative to clinician labels. At a 91% specificity, 94% of thick NT cases were detected (sensitivity) in the test set. The detection rate was statistically higher (p = 0.003) in high versus low-quality images (AUROC 0.98 vs. 0.90, respectively). CONCLUSION: Our model provides an explainable deep-learning method for detecting increased NT. This technique can be used to screen for other fetal anomalies in the first trimester of pregnancy.

A simulation study to assess the potential benefits of MRI-based fetal weight estimation as a second-line test for suspected macrosomia.

OBJECTIVE: To simulate the outcomes of Boulvain's trial by using magnetic resonance imaging (MRI) for estimated fetal weight (EFW) as a second-line confirmatory imaging. STUDY DESIGN: Data derived from the Boulvain's trial and the study PREMACRO (PREdict MACROsomia) were used to simulate a 1000-patient trial. Boulvain's trial compared induction of labor (IOL) to expectant management in suspected macrosomia, whereas PREMACRO study compared the performance of ultrasound-EFW (US-EFW) and MRI-EFW in the prediction of birthweight. The primary outcome was the incidence of significant shoulder dystocia (SD). Cesarean delivery (CD), hyperbilirubinemia (HB), and IOL at < 39 weeks of gestation (WG) were selected as secondary outcomes. A subgroup analysis of the Boulvain's trial was performed to estimate the incidence of the primary and secondary outcomes in the true positive and false positive groups for the two study arms. Sensitivity, specificity, positive and negative predictive values (PPV, NPV) for the prediction of macrosomia by MRI-EFW at 36 WG were calculated, and a decision tree was constructed for each outcome. RESULTS: The PPV of US-EFW for the prediction of macrosomia in the PREMACRO trial was 56.3 %. MRI-EFW was superior to US-EFW as a predictive tool resulting in lower rates of induction for false-positive cases. Repeating Boulvain's trial using MRI-EFW as a second-line test would result in similar rates of SD (relative risk [RR]:0.36), CD (RR:0.84), and neonatal HB (RR:2.6), as in the original trial. Increasing the sensitivity and specificity of MRI-EFW resulted in a similar relative risk for SD as in Boulvain's trial, but with reduced rates of IOL < 39 WG, and improved the RR of CD in favor of IOL. We found an inverse relationship between IOL rate and incidence of SD for both US-EFW and MRI-EFW, although overall rates of IOL, CD, and neonatal HB would be lower with MRI-derived estimates of fetal weight. CONCLUSION: The superior accuracy of MRI-EFW over US-EFW for the diagnosis of macrosomia could result in lower rates of IOL without compromising the relative advantages of the intervention but fails to demonstrate a significant benefit to justify a replication of the original trial using MRI-EFW as a second-line test.

Indicated vs universal third-trimester ultrasound examination in low-risk pregnancies: a pre-post-intervention study.

BACKGROUND: In low-risk pregnancies, a third-trimester ultrasound examination is indicated if fundal height measurement and gestational age discrepancy are observed. Despite potential improvement in the detection of ultrasound abnormality, prior trials to date on universal third-trimester ultrasound examination in low-risk pregnancies, compared with indicated ultrasound examination, have not demonstrated improvement in neonatal or maternal adverse outcomes. OBJECTIVE: The primary objective was to determine if universal third-trimester ultrasound examination in low-risk pregnancies could attenuate composite neonatal adverse outcomes. The secondary objectives were to compare changes in composite maternal adverse outcomes and detection of abnormalities of fetal growth (fetal growth restriction or large for gestational age) or amniotic fluid (oligohydramnios or polyhydramnios). STUDY DESIGN: Our pre-post intervention study at 9 locations included low-risk pregnancies, those without indication for ultrasound examination in the third trimester. Compared with indicated ultrasound in the preimplementation period, in the postimplementation period, all patients were scheduled for ultrasound examination at 36.0-37.6 weeks. In both periods, clinicians intervened on the basis of abnormalities identified. Composite neonatal adverse outcomes included any of: Apgar score ≤5 at 5 minutes, cord pH <7.00, birth trauma (bone fracture or brachial plexus palsy), intubation for >24 hours, hypoxic-ischemic encephalopathy, seizure, sepsis (bacteremia proven with blood culture), meconium aspiration syndrome, intraventricular hemorrhage grade III or IV, periventricular leukomalacia, necrotizing enterocolitis, stillbirth after 36 weeks, or neonatal death within 28 days of birth. Composite maternal adverse outcomes included any of the following: chorioamnionitis, wound infection, estimated blood loss >1000 mL, blood transfusion, deep venous thrombus or pulmonary embolism, admission to intensive care unit, or death. Using Bayesian statistics, we calculated a sample size of 600 individuals in each arm to detect >75% probability of any reduction in primary outcome (80% power; 50% hypothesized risk reduction). RESULTS: During the preintervention phase, 747 individuals were identified during the initial ultrasound examination, and among them, 568 (76.0%) met the inclusion criteria at 36.0-37.6 weeks; during the postintervention period, the corresponding numbers were 770 and 661 (85.8%). The rate of identified abnormalities of fetal growth or amniotic fluid increased from between the pre-post intervention period (7.1% vs 22.2%; P<.0001; number needed to diagnose, 7; 95% confidence interval, 5-9). The primary outcome occurred in 15 of 568 (2.6%) individuals in the preintervention and 12 of 661 (1.8%) in the postintervention group (83% probability of risk reduction; posterior relative risk, 0.69 [95% credible interval, 0.34-1.42]). The composite maternal adverse outcomes occurred in 8.6% in the preintervention and 6.5% in the postintervention group (90% probability of risk; posterior relative risk, 0.74 [95% credible interval, 0.49-1.15]). The number needed to treat to reduce composite neonatal adverse outcomes was 121 (95% confidence interval, 40-200). In addition, the number to reduce composite maternal adverse outcomes was 46 (95% confidence interval, 19-74), whereas the number to prevent cesarean delivery was 18 (95% confidence interval, 9-31). CONCLUSION: Among low-risk pregnancies, compared with routine care with indicated ultrasound examination, implementation of a universal third-trimester ultrasound examination at 36.0-37.6 weeks attenuated composite neonatal and maternal adverse outcomes.

Echogenic intracardiac foci detection and location in the second-trimester ultrasound and association with fetal outcomes: A systematic literature review.

BACKGROUND: Echogenic Intracardiac Foci (EIF) are non-structural markers identified during the routine 18-20-week foetal anomaly ultrasound scan yet their clinical significance on future outcomes for the infant is unclear. OBJECTIVE: To examine the association between EIF and risk of preterm birth, chromosomal abnormalities, and cardiac abnormalities. DESIGN: A review across four databases to identify English language journal articles of EIF using a cohort study design. All studies were reviewed for quality using the Critical Appraisal Skills Programme (CASP) checklist and data extracted for comparison and analysis. RESULTS: 19 papers from 9 different countries were included. Combining these studies showed 4.6% (95% CI = 4.55-4.65%) of all pregnancies had EIF which was on the left in 86% of cases, on the right in 3% of cases and bilaterally in 10%. There was no evidence that EIF was associated with higher rates of preterm birth. However, it is possible that infants with EIF were more likely to be terminated rather than be born preterm as there was a 2.1% (range 0.3-4.2%) rate of termination or death of the foetus after week 20 among those with EIF. There was no evidence that EIF alone is highly predictive of chromosomal abnormalities. There was evidence that EIF is associated with higher rates of minor cardiac abnormalities (e.g. ventricular septal defect, tricuspid regurgitation or mitral regurgitation)) with 5.1% (224 of 4385) of those with EIF showing cardiac abnormalities (3.08% in retrospective studies and 17.85% in prospective studies). However, the risk of cardiac defects was only higher with right-sided EIF and where the EIF persisted into the third trimester. However, this is a rare event and would be seen in an estimated 4 per 10,000 pregnancies. CONCLUSION: EIF alone was not associated with adverse outcomes for the infant. Only persistent EIF on the right side showed evidence of carrying a higher risk of cardiac abnormality and would warrant further follow-up.

A deep learning framework for identifying and segmenting three vessels in fetal heart ultrasound images.

BACKGROUND: Congenital heart disease (CHD) is one of the most common birth defects in the world. It is the leading cause of infant mortality, necessitating an early diagnosis for timely intervention. Prenatal screening using ultrasound is the primary method for CHD detection. However, its effectiveness is heavily reliant on the expertise of physicians, leading to subjective interpretations and potential underdiagnosis. Therefore, a method for automatic analysis of fetal cardiac ultrasound images is highly desired to assist an objective and effective CHD diagnosis. METHOD: In this study, we propose a deep learning-based framework for the identification and segmentation of the three vessels-the pulmonary artery, aorta, and superior vena cava-in the ultrasound three vessel view (3VV) of the fetal heart. In the first stage of the framework, the object detection model Yolov5 is employed to identify the three vessels and localize the Region of Interest (ROI) within the original full-sized ultrasound images. Subsequently, a modified Deeplabv3 equipped with our novel AMFF (Attentional Multi-scale Feature Fusion) module is applied in the second stage to segment the three vessels within the cropped ROI images. RESULTS: We evaluated our method with a dataset consisting of 511 fetal heart 3VV images. Compared to existing models, our framework exhibits superior performance in the segmentation of all the three vessels, demonstrating the Dice coefficients of 85.55%, 89.12%, and 77.54% for PA, Ao and SVC respectively. CONCLUSIONS: Our experimental results show that our proposed framework can automatically and accurately detect and segment the three vessels in fetal heart 3VV images. This method has the potential to assist sonographers in enhancing the precision of vessel assessment during fetal heart examinations.

IMU-Assisted Manual 3D-Ultrasound Imaging Using Motion-Constrained Swept-Fan Scans.

Three-dimensional (3D) ultrasonic imaging can enable post-facto plane of interest selection. It can be performed with devices such as wobbler probes, matrix probes, and sensor-based probes. Ultrasound systems that support 3D-imaging are expensive with added hardware complexity compared to 2D-imaging systems. An inertial measurement unit (IMU) can potentially be used for 3D-imaging by using it to track the motion of a one-dimensional array probe and constraining its motion in one degree of freedom (1-DoF) rotation (swept-fan). This work demonstrates the feasibility of an affordable IMU-assisted manual 3D-ultrasound scanner (IAM3US). A consumer-grade IMU-assisted 3D scanner prototype is designed with two support structures for swept-fan. After proper IMU calibration, an appropriate KF-based algorithm estimates the probe orientation during the swept-fan. An improved scanline-based reconstruction method is used for volume reconstruction. The evaluation of the IAM3US system is done by imaging a tennis ball filled with water and the head region of a fetal phantom. From fetal phantom reconstructed volumes, suitable 2D planes are extracted for biparietal diameter (BPD) manual measurements. Later, in-vivo data is collected. The novel contributions of this paper are (1) the application of a recently proposed algorithm for orientation estimation of swept-fan for 3D imaging, chosen based on the noise characteristics of selected consumer grade IMU (2) assessment of the quality of the 1-DoF swept-fan scan with a deflection detector along with monitoring of maximum angular rate during the scan and (3) two probe holder designs to aid the operator in performing the 1-DoF rotational motion and (4) end-to-end 3D-imaging system-integration. Phantom studies and preliminary in-vivo obstetric scans performed on two patients illustrate the usability of the system for diagnosis purposes.