RESUMEN

PURPOSE: This study aims to investigate the fetal Evans Index and establish a nomogram for fetuses without any additional fetal anomalies detected during the prenatal period. METHODS: We conducted our research at Ankara City Hospital, including 894 patients who were admitted and evaluated between gestational weeks 16-40. These patients had no fetal anomalies detected in subsequent gestational weeks. Descriptive data, such as age, gravidity, parity, and body mass index (BMI), were recorded. Gestational week and Evans Index (mean, median, standard deviation, minimum, maximum, and percentile) were also documented. The Evans index was calculated as the ratio between the maximal width of the frontal horns and the maximal width of the inner diameter of the cranium. RESULTS: We evaluated 894 fetuses in pregnant women had no fetal anomalies detected throughout the pregnancy. The evaluation took place at different gestational weeks, and a nomogram for the Evans Index was created. CONCLUSIONS: It is relevant for clinicians and researchers to be aware of the range of fetal Evans Index values across different gestational weeks as a prognostic criterion.

RESUMEN

OBJECTIVES: The aim of this study is to establish the normal calcaneus-metatarsal angles in order to facilitate the recognition of fetal foot deformities. Therefore, a novel measurement technique was determined using ultrasound and applied to nonanomalous fetuses in the 18th to 23rd gestational weeks. MATERIALS AND METHODS: This cross-sectional study included 100 low-risk, nonanomalous fetuses in the 18th to 23rd weeks of gestation. Measurements were obtained using the novel measurement technique in fetal ultrasound examination. Fetal biometry, foot sole length, first and fifth metatarsal diaphyseal length measurements were taken. The acute angle measurement of the line segment passing between the calcaneus outer lateral and proximal fifth metatarsal with the long axis of the first metatarsal diaphysis (CA-mtt1) and long axis of the fifth metatarsal diaphysis (CA-mtt5) was taken. Interobserver and intraobserver reliabilities were assessed by intraclass correlations (ICC). RESULTS: A strong positive correlation was detected between bi-parietal diameter (BPD) and foot sole length, first metatarsal diaphyseal length and fifth metatarsal diaphyseal length (p < 0.0001). No significant correlation of these CA-mtt1 and CA-mtt5 angles with BPD was detected (p = 0.35, p = 0.82, respectively). The data suggest that the CA-mtt1 and CA-mtt5 angles remained consistently within a narrow range and were determined to be 20.0° ± 8.7° and 7.8° ± 7.5°, respectively. Intraobserver and interobserver agreement for CA-mtt1 angle was moderate (ICC, 0.655) and moderate (ICC, 0.615), for CA-mtt5 angle was moderate (ICC, 0.631) and moderate (ICC, 0.605), respectively. CONCLUSION: A normal reference range was established for fetal sole length, first metatarsal diaphysis length, fifth metatarsal diaphysis length, CA-mtt1 angle, and CA-mtt5 angle in the 18th to 23rd weeks of gestation. It was determined that the CA-mtt1 and CA-mtt5 angles remained relatively constant within a narrow range throughout the assessed gestational weeks.

RESUMEN

Background/Objectives: Non-Invasive prenatal test (NIPT) is used as a universal or contingent test after prior risk assessment. Screening is mainly performed for common trisomies (T21, T13, T18), although other chromosomal anomalies may be detected. Our objective was to study the performance of GWNIPT in the detection of chromosomal abnormalities in pregnancies in which an invasive prenatal study was performed and in early pregnancy losses, in comparison with the reference test. Method: VeriSeqTM NIPT Solution v2, a genome-wide NIPT (GWNIPT), was performed prior to invasive testing in fetal diagnostic study cases (FDS, n = 155) and in early pregnancy losses (EPL, n = 68). Results: In the FDS group, the diagnostic test (QFPCR, array and karyotype) detected anomalies in 32 pregnancies (21%), in twenty of them (61%) also detected by GWNIPT. Eleven of the twelve cases undetected by GWNIPT were balanced translocations (n = 4) or deletions/duplications <7 Mb (n = 7). In the EPL group, GWNIPT detected anomalies in 46% of cases (31/68) but comparison with reference test (QFPCR and karyotype) in products of conception (POC) was only possible in 18 cases. Concordant results between POC and GWNIPT test were obtained in 16 of the 18 cases. In EPL, with GWNIPT testing, common trisomies accounted for 25.8% of cases (8/31), rare trisomies 54.8% (17/31) and microdeletions/duplications 16.1% (5/31). Conclusions: The GWNIPT test may be useful in clinical practice in prenatal and in EPL's genetic diagnosis when the appropriate sample is not available.

RESUMEN

BACKGROUND: The impacts of maternal depression during mid-to-late pregnancy on fetal growth have been extensively investigated. However, the association between maternal depression during early pregnancy and fetal intrauterine growth are less clear. METHODS: A prospective study comprised 23,465 eligible pregnant women and their offspring was conducted at a hospital-based center in Shanghai. Prenatal depression was assessed used using Patient Health Questionnaire (PHQ-9) before 14 gestational weeks. Differences in fetal growth trajectory of different maternal depressive statuses during three periods (16-23, 24-31, and 32-41 gestational weeks) were compared using a multilevel model with fractional polynomials. RESULTS: Women with depressive symptoms during early pregnancy had higher longitudinal fetal trajectories, with an estimated increase in fetal weight (ß = 0.33; 95 % CI, 0.06-0.61), compared to those without depressive symptoms. Increases in fetal abdominal circumference among women with depressive symptoms were observed before 23 gestational weeks. Offspring born to mothers with early pregnancy depression had a significantly higher birth weight of 14.13 g (95 % CI, 1.33-27.81 g) and an increased risk of severe large size for gestational age (adjusted odds ratio [aOR], 1.64; 95 % CI, 1.32-2.04) and macrosomia (aOR, 1.21; 95 % CI, 1.02-1.43). LIMITATIONS: Self-rated scale was used to assess depressive symptoms rather than clinical diagnosis. And Long-term effects of early pregnancy depression on offspring were not explored. CONCLUSIONS: The study revealed an association between maternal depression during early pregnancy and increased fetal biometrics, higher birth weight, and an elevated risk of severe large size for gestational age and macrosomia.

Asunto(s)

RESUMEN

We describe two cases of umbilical cord (UC) angiomyxoma diagnosed prenatally by sonography in the second trimester of pregnancy. In both cases, a complex mass was detected at the placental insertion site, characterized by an echoic nodule surrounding the umbilical vessels and distal edematous Wharton's jelly. Follow-up scans showed that the mass grew mainly at the expense of its edematous component, with normal uteroplacental Dopplers throughout the remaining of the pregnancy. However, late-onset fetal growth restriction complicated the progress of pregnancy, requiring delivery by Cesarean section at 37 weeks' gestation in both cases. Neonatal courses were unremarkable. An extensive review of the English literature was also performed, collecting 45 similar cases including ours. Our experience as well as the review of the literature confirms that UC angiomyxoma is an uncommon, sporadic condition that is usually detected incidentally during prenatal sonography and presents as an isolated finding. Nevertheless, it represents a high-risk condition for pregnancy complications including prematurity, fetal growth restriction, and fetal demise.

Asunto(s)

RESUMEN

Over the past decade, deep-learning (DL) algorithms have become a promising tool to aid clinicians in identifying fetal head standard planes (FHSPs) during ultrasound (US) examination. However, the adoption of these algorithms in clinical settings is still hindered by the lack of large annotated datasets. To overcome this barrier, we introduce FetalBrainAwareNet, an innovative framework designed to synthesize anatomically accurate images of FHSPs. FetalBrainAwareNet introduces a cutting-edge approach that utilizes class activation maps as a prior in its conditional adversarial training process. This approach fosters the presence of the specific anatomical landmarks in the synthesized images. Additionally, we investigate specialized regularization terms within the adversarial training loss function to control the morphology of the fetal skull and foster the differentiation between the standard planes, ensuring that the synthetic images faithfully represent real US scans in both structure and overall appearance. The versatility of our FetalBrainAwareNet framework is highlighted by its ability to generate high-quality images of three predominant FHSPs using a singular, integrated framework. Quantitative (Fréchet inception distance of 88.52) and qualitative (t-SNE) results suggest that our framework generates US images with greater variability compared to state-of-the-art methods. By using the synthetic images generated with our framework, we increase the accuracy of FHSP classifiers by 3.2% compared to training the same classifiers solely with real acquisitions. These achievements suggest that using our synthetic images to increase the training set could provide benefits to enhance the performance of DL algorithms for FHSPs classification that could be integrated in real clinical scenarios.

Asunto(s)

RESUMEN

In obstetric ultrasound (US) scanning, the learner's ability to mentally build a three-dimensional (3D) map of the fetus from a two-dimensional (2D) US image represents a significant challenge in skill acquisition. We aim to build a US plane localization system for 3D visualization, training, and guidance without integrating additional sensors. This work builds on top of our previous work, which predicts the six-dimensional (6D) pose of arbitrarily oriented US planes slicing the fetal brain with respect to a normalized reference frame using a convolutional neural network (CNN) regression network. Here, we analyze in detail the assumptions of the normalized fetal brain reference frame and quantify its accuracy with respect to the acquisition of transventricular (TV) standard plane (SP) for fetal biometry. We investigate the impact of registration quality in the training and testing data and its subsequent effect on trained models. Finally, we introduce data augmentations and larger training sets that improve the results of our previous work, achieving median errors of 2.97 mm and 6.63° for translation and rotation, respectively.

RESUMEN

INTRODUCTION: This study aims to investigate probe motion during full mid-trimester anomaly scans. METHODS: We undertook a prospective, observational study of obstetric sonographers at a UK University Teaching Hospital. We collected prospectively full-length video recordings of routine second-trimester anomaly scans synchronized with probe trajectory tracking data during the scan. Videos were reviewed and trajectories analyzed using duration, path metrics (path length, velocity, acceleration, jerk, and volume) and angular metrics (spectral arc, angular area, angular velocity, angular acceleration, and angular jerk). These trajectories were then compared according to the participant level of expertise, fetal presentation, and patient BMI. RESULTS: A total of 17 anomaly scans were recorded. The average velocity of the probe was 12.9 ± 3.4 mm/s for the consultants versus 24.6 ± 5.7 mm/s for the fellows (p = 0.02), the average acceleration 170.4 ± 26.3 mm/s2 versus 328.9 ± 62.7 mm/s2 (p = 0.02), and the average jerk 7491.7 ± 1056.1 mm/s3 versus 14944.1 ± 3146.3 mm/s3 (p = 0.02), the working volume 9.106 ± 4.106 mm3 versus 29.106 ± 11.106 mm3 (p = 0.03), respectively. The angular metrics were not significantly different according to the participant level of expertise, the fetal presentation, or to patients BMI. CONCLUSION: Some differences in the probe path metrics (velocity, acceleration, jerk and working volume) were noticed according to operator's level.

Asunto(s)

RESUMEN

Goal: FetSAM represents a cutting-edge deep learning model aimed at revolutionizing fetal head ultrasound segmentation, thereby elevating prenatal diagnostic precision. Methods: Utilizing a comprehensive dataset-the largest to date for fetal head metrics-FetSAM incorporates prompt-based learning. It distinguishes itself with a dual loss mechanism, combining Weighted DiceLoss and Weighted Lovasz Loss, optimized through AdamW and underscored by class weight adjustments for better segmentation balance. Performance benchmarks against prominent models such as U-Net, DeepLabV3, and Segformer highlight its efficacy. Results: FetSAM delivers unparalleled segmentation accuracy, demonstrated by a DSC of 0.90117, HD of 1.86484, and ASD of 0.46645. Conclusion: FetSAM sets a new benchmark in AI-enhanced prenatal ultrasound analysis, providing a robust, precise tool for clinical applications and pushing the envelope of prenatal care with its groundbreaking dataset and segmentation capabilities.

RESUMEN

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.

Asunto(s)

RESUMEN

BACKGROUND: Maternal obesity influences birth weight and newborn adiposity. Fetal fractional limb volume has recently been introduced as a useful parameter for the proxy of fetal adiposity. However, the association between maternal adiposity and the growth of fetal fractional limb volume has not been examined. AIMS: To investigate the association of maternal pre-pregnancy BMI with the growth of fetal fractional limb volume. STUDY DESIGN: Prospective cohort study. SUBJECTS: Women with singleton uncomplicated pregnancies enrolled between July 2017 and June 2020. OUTCOME MEASURES: Fetal fractional limb volume was assessed between 20 and 40 weeks' gestation, measured as cylindrical limb volume based on 50 % of the total diaphysis length. The measured fractional limb volume at each gestational week were converted to z-scores based on a previous report. The association between pre-pregnancy BMI and fetal fractional limb volume was examined. Maternal age, parity, gestational weight gain and fetal sex were considered as potential confounding variables. RESULTS: Ultrasound scans of 455 fractional arm volume and thigh volume were obtained. Fractional limb volume increased linearly until the second trimester of gestation, then increased exponentially in the third trimester. Maternal pre-pregnancy BMI was significantly correlated with z-scores of fractional arm volume and thigh volume across gestation. The post-hoc analysis showed the association between pre-pregnancy BMI and fractional arm volume was significant especially between 34 and 40 weeks. CONCLUSIONS: Maternal obesity influences the growth pattern of fetal fractional limb volume. Fractional arm volume may potentially provide a useful surrogate marker of fetal nutritional status in late gestation.

Asunto(s)

RESUMEN

This work measures temporal signal-to-noise ratio (SNR) thresholds that indicate when random noise during ultrasound scanning becomes imperceptible to expert human observers. Visible noise compromises image quality and can potentially lead to non-diagnostic scans. Noise can arise from both stable acoustic sources (clutter) or randomly varying electronic sources (temporal noise). Extensive engineering effort has focused on decreasing noise in both of these categories. In this work, an observer study with five practicing sonographers was performed to assess sonographer sensitivity to temporal noise in ultrasound cine clips. Understanding the conditions where temporal noise is no longer visible during ultrasound imaging can inform engineering efforts seeking to minimize the impact this noise has on image quality. The sonographers were presented with paired temporal noise-free and noise-added simulated speckle cine clips and asked to select the noise-added clips. The degree of motion in the imaging target was found to have a significant effect on the SNR levels where noise was perceived, while changing imaging frequency had little impact. At realistic in vivo motion levels, temporal noise was not perceived in cine clips at and above 28 dB SNR. In a case study presented here, the potential of adaptive intensity adjustment based on this noise perception threshold is validated in a fetal imaging scenario. This study demonstrates how noise perception thresholds can be applied to help design or tune ultrasound systems for different imaging tasks and noise conditions.

Asunto(s)

RESUMEN

BACKGROUND: Pregnancy loss occurring before 20 weeks gestation is referred to as miscarriage. Various clinical presentations of miscarriage include threatened, inevitable, incomplete, complete, septic, and missed miscarriage. Early-stage threatened miscarriage may manifest with symptoms such as abdominal discomfort and vaginal bleeding. Threatened miscarriage is clinically defined as the manifestation of positive fetal heart sounds in pregnancies occurring before the 20th week of gestation, concomitant with vaginal bleeding and a closed cervix. OBJECTIVES: The primary aim of this study was to evaluate the association between serum C-reactive protein (CRP) levels and fetal ultrasound findings in the prediction of threatened miscarriage during the first trimester of pregnancy. METHODS: In this prospective case-control study, a total of 100 pregnant women at 7-13 weeks of gestation were enrolled. All participants initially presented with a singleton embryo displaying cardiac activity on ultrasound. The study cohort was divided into two groups: Group 1 consisted of 50 women with uncomplicated pregnancies, while Group 2 comprised 50 women experiencing symptoms indicative of threatened miscarriage. RESULTS: Notably, within Group 2, patients who eventually experienced miscarriage exhibited significantly elevated serum high-sensitivity CRP levels in comparison to those who maintained their pregnancies. CONCLUSIONS: Threatened miscarriage cases demonstrated a substantial increase in serum high-sensitivity CRP levels compared to the control group. Furthermore, CRP levels exhibited a correlation with the risk of miscarriage, suggesting their potential utility in conjunction with ultrasound parameters for prognosticating threatened miscarriage during the first trimester.

RESUMEN

BACKGROUND: According to prenatal ultrasonographic studies, single umbilical artery may be present alone or in association with other fetal abnormalities. So far, the exact pathogenesis of bladder exstrophy is unclear. Some scholars believe that bladder exstrophy and cloacal exstrophy should be regarded as a disease spectrum to explore their pathogenesis. If bladder exstrophy and cloacal exstrophy are regarded as the same disease spectrum, then we can speculate that the single umbilical artery should have the probability of being accompanied by bladder exstrophy at the same time. CASE PRESENTATION: For the first time, we report a rare case of fetal bladder exstrophy with single umbilical artery in single pregnancy. This patient underwent targeted color Doppler ultrasound at 26 weeks of pregnancy which first suspected bladder exstrophy with single umbilical artery and fetal MRI for diagnosis at 38 + 3 weeks of pregnancy which confirmed the suspicion. After the diagnosis was confirmed, the patient was scheduled for a multidisciplinary discussion. Ultimately the patient opted for induced fetal demise at 38 + 5 weeks of pregnancy and the physical appearance of the fetal demise affirmed previous ultrasound and MRI examination results. CONCLUSIONS: Our report is the first finding of single umbilical artery combined with bladder exstrophy in a singleton pregnancy. Accordingly, our case enhances the evidence that cloacal exstrophy and bladder exstrophy should be treated as the same disease spectrum. In addition, we conducted a literature review on the diagnostic progress of single umbilical artery combined with bladder exstrophy, hoping to provide useful references for the diagnosis of this disease.

Asunto(s)

RESUMEN

OBJECTIVE: The aim of this study was address the challenges posed by the manual labeling of fetal ultrasound images by introducing an unsupervised approach, the fetal ultrasound semantic clustering (FUSC) method. The primary objective was to automatically cluster a large volume of ultrasound images into various fetal views, reducing or eliminating the need for labor-intensive manual labeling. METHODS: The FUSC method was developed by using a substantial data set comprising 88,063 images. The methodology involves an unsupervised clustering approach to categorize ultrasound images into diverse fetal views. The method's effectiveness was further evaluated on an additional, unseen data set consisting of 8187 images. The evaluation included assessment of the clustering purity, and the entire process is detailed to provide insights into the method's performance. RESULTS: The FUSC method exhibited notable success, achieving >92% clustering purity on the evaluation data set of 8187 images. The results signify the feasibility of automatically clustering fetal ultrasound images without relying on manual labeling. The study showcases the potential of this approach in handling a large volume of ultrasound scans encountered in clinical practice, with implications for improving efficiency and accuracy in fetal ultrasound imaging. CONCLUSION: The findings of this investigation suggest that the FUSC method holds significant promise for the field of fetal ultrasound imaging. By automating the clustering of ultrasound images, this approach has the potential to reduce the manual labeling burden, making the process more efficient. The results pave the way for advanced automated labeling solutions, contributing to the enhancement of clinical practices in fetal ultrasound imaging. Our code is available at https://github.com/BioMedIA-MBZUAI/FUSC.

Asunto(s)

RESUMEN

BACKGROUND: Early morphologic ultrasound, generally carried out in case of atypical first trimester serum markers (PAPP-A and/or free hCGß <0.30 MoM), has not been re-evaluated since the possibility of performing a cell-free fetal DNA analysis in this indication. Our objective was to evaluate the usefulness of early morphological ultrasound in case of atypical profile of serum markers performed in association with Non-Invasive Prenatal Testing (NIPT). METHODS: This was a single-center retrospective study in a tertiary maternity. Between January 2017 and December 2021, women with an atypical first trimester serum markers and low/intermediate risk for trisomy 21 (<1/50) were included. The clinical data, results of first trimester serum markers, NIPT, early morphological ultrasound and subsequent ultrasounds and other investigations (amniocentesis, pregnancy outcomes) were analyzed. RESULTS: After exclusion of women with high-risk of trisomy 21 and lost to follow-up, 163 women were included. In 72 % of cases (117/163), women had a low risk of trisomy 21, and 39 % (59/163) had an early morphological ultrasound. Early morphological ultrasound was useful to detect severe IUGR leading to the suspicion of triploidy (3/163, 1.8 %). In all other situations, it did not allow earlier management. After analysis of the 3 triploidy cases, a collapsed profile for both serum markers was demonstrated (<0.25 MoM). CONCLUSIONS: Systematic early morphological ultrasound in case of an atypical serum marker profile seems useless considering the performance of NIPT. An ultrasound restricted to women with both markers below 0.25 MoM would allow the early detection of triploidy.

Asunto(s)

RESUMEN

OBJECTIVES: In the first trimester, intertwin crown-rump length (CRL) discordance has emerged as a notable factor linked to adverse perinatal health effects. It is frequently employed as a basis for counseling parents regarding potential adverse pregnancy outcomes. Despite its established association with adverse outcomes, the significance of CRL discordance in substantially predicting pregnancy problems and its efficacy in pregnancy screening remain subjects of ongoing discussion. The aim of this manuscript is to present current knowledge on CRL discordance. METHODS: PubMed was searched for related articles with terms "Crown-Rump length", "Prenatal Screening", "Twin pregnancy", "Discordance". RESULTS: Twenty-two studies were included in our study with six reporting data on monochorionic and 16 assessing the correlation between CRL discordance and adverse pregnancy outcomes. Fetal loss at the 20th and 24th week of the pregnancy, SGA neonates, pre-term delivery (32 weeks), perinatal death (24 weeks) are all reported adverse outcomes associated with CRL discordance. The reported cut-off for increased risk of adverse perinatal outcomes is a discordance of at least 10% or more. CONCLUSIONS: Increased CRL (>10 %) discordance is linked to a higher risk of sFGR in both monochorionic and dichorionic pregnancies, fetal loss, and preterm delivery.

Asunto(s)

RESUMEN

OBJECTIVE: To evaluate perinatal and postnatal outcomes of fetuses with an isolated small head circumference (HC) on expert ultrasound examination in the second trimester for further recommendations in prenatal care. STUDY DESIGN: In a retrospective cohort we included singleton-pregnancies with a fetal HC > -3.0 SD and ≤ -1.64 SD determined on expert ultrasound examination between 18 and 24 weeks of gestational age. Three subgroups were determined: "isolated small HC (ISHC)", "small HC plus abdominal circumference (AC) ≤ p10 (SHC+)" and "small HC plus AC ≤ p10 and Doppler abnormalities (SHC + D)". After ultrasound examination, genetic testing was sometimes offered and postnatally genetic tests were performed on indication. RESULTS: We included 252 pregnancies: 109 ISHC, 104 SHC+, and 39 SHC + D. In the ISHC and SHC+ subgroup, 96 % of the fetuses were born alive and did not die neonatal. In the SH + D group this was only 38 %. In the SHC+ subgroup, less fetuses were delivered vaginal (non-instrumental) compared to the ISHC subgroup (61 % vs. 73 %, p < 0.01). In the ISHC and SHC+ subgroup s some fetuses were diagnosed with congenital defects (4 % vs. 10 %, p = 0.08) and with a genetic anomaly (6.4 % vs. 7.7 %, p = 0.13) after 24 weeks or postnatally. In SHC + D subgroups 5 % presented with congenital defects and 2.6 % with a genetic anomaly. CONCLUSION: We conclude that fetuses with a small HC without structural anomalies on second trimester expert ultrasound require follow-up and special medical attention. We recommend differentiating between ISHC, SHC+, and SHC + D for prenatal counseling. Genetic testing and referral to a clinical geneticist should be considered.

Asunto(s)

RESUMEN

In clinical practice, ultrasound standard planes (SPs) selection is experience-dependent and it suffers from inter-observer and intra-observer variability. Automatic recognition of SPs can help improve the quality of examinations and make the evaluations more objective. In this paper, we propose a method for the automatic identification of SPs, to be installed onboard a portable ultrasound system with limited computational power. The deep Learning methodology we design is based on the concept of Knowledge Distillation, transferring knowledge from a large and well-performing teacher to a smaller student architecture. To this purpose, we evaluate a set of different potential teachers and students, as well as alternative knowledge distillation techniques, to balance a trade-off between performances and architectural complexity. We report a thorough analysis of fetal ultrasound data, focusing on a benchmark dataset, to the best of our knowledge the only one available to date.

Asunto(s)

RESUMEN

OBJECTIVES: This retrospective study aimed to assess disparities between prenatal ultrasound and autopsy findings in pregnancies that resulted in fetal loss, and to evaluate the diagnostic performance of prenatal ultrasound using postmortem examinations as a gold standard. METHODS: Our study included 136 autopsy cases following a fetal loss that occurred at our tertiary medical center for 8 years. A comparison between the prenatal ultrasound and autopsy findings was made, and all cases were classified according to the degree of agreement. The diagnostic performance of prenatal ultrasound was calculated at the level of organ system and specific malformations. RESULTS: The primary sonographic diagnosis was confirmed in 91.9% of the cases (n = 125). General agreement was highest among central nervous system (CNS), cardiovascular and musculoskeletal systems (85.7%, n = 36, 18, and 12, respectively) and lowest among facial, multiple anomalies, genitourinary and gastrointestinal systems (50.0%, 74.3%, 78.6%, and 80.0%, n = 2, 26, 11, and 4, respectively). The sensitivity of ultrasound was highest in the CNS (93.2%) and musculoskeletal (87.0%) and lowest in the facial (32.3%) and pulmonary (13.0%) systems. Specifically, low diagnostic rates were noted in detecting ventriculomegaly, valvular anomalies, renal dysplasia, spleen and adrenal anomalies, and digital and facial defects. CONCLUSIONS: Our study observed an overall high agreement between prenatal ultrasound and autopsy while contributing to our comprehensive understanding of its strengths and limitations across various types of organ systems and specific malformations.

Asunto(s)