Constructive learning: a high-performance framework for fetal head circumference estimation

Background: Labor is the physiological process by which the fetus and placenta are expelled from the uterus through the vaginal canal. Labor is influenced by maternal effort, uterine contractions, fetal characteristics, and pelvic anatomy. Management of normal labor involves monitoring maternal vitals, cervical progress, and labs, while minimizing interventions. Active labor, once defined at 4 cm dilation, is now considered to begin at 6 cm, impacting obstetric management and outcomes. Objectives: This study aims to compare maternal and neonatal outcomes, alongside labor interventions, when defining active labor onset at 4 cm versus 6 cm cervical dilation. Methods: A prospective case-control study was conducted at Al-Mawanee Teaching Hospital, Basrah, for the period from 1st of November 2024 to 30th of July 2025. Compared adverse obstetric outcomes in low-risk women admitted at 4 cm versus 6 cm cervical dilation. Eligible term, singleton, cephalic pregnancies were included, excluding medical disorders, fetal complications, inductions, and prior caesarean section. Data collection involved questionnaires, examinations, and monitoring of labor management, maternal complications, and neonatal outcomes. Results: This study compared outcomes among 200 women admitted in labor at 4 cm versus 6 cm cervical dilation. Significant differences included maternal age (older in 6 cm group, p=0.04) and parity (more nulliparas at 4 cm, p=0.003). Cervical consistency was softer in the 6 cm group (p=0.028), and fetal head station was more advanced (p=0.05). Women admitted at 4 cm had longer labor duration, longer amniotomy-to-delivery time, and higher oxytocin augmentation use (all p<0.001). Cesarean indications differed (p=0.032): fetal distress predominated at 6 cm, poor progress at 4 cm. maternal complications and neonatal outcomes showed no significant differences. Conclusion: Admission at 4 cm was linked to longer labor, more oxytocin use, and caesareans for poor progress, while 6 cm showed better readiness. Overall caesarean rates, maternal complications, and neonatal outcomes were similar.

ABSTRACTObjectiveTo quantitatively evaluate bladder morphological changes induced by fetal head descent during late pregnancy using transperineal ultrasonography (US) and to investigate the association with urinary incontinence (UI). This study aimed to introduce a novel, imaging‐based approach for assessing pregnancy‐related urinary dysfunction.MethodsIn this study, 14 women with singleton pregnancies beyond 36 weeks of gestation were evaluated. During routine antenatal visits, participants completed a validated questionnaire assessing urinary symptoms, and the bladder descent angle (BDA) was measured using transperineal US. The BDA was defined as the angle between the bladder base and the inferior margin of the pubic symphysis.ResultsUI was reported in 57.1% of participants, with stress urinary incontinence (SUI) being the most common subtype (62.5%). Ultrasonography revealed that BDA increased in late pregnancy. The SUI group exhibited a significantly greater BDA compared with the no‐UI group (p = 0.03), whereas the overall UI group showed only a non‐significant trend (p = 0.081).ConclusionsThe BDA assessed by transperineal US is considered a practical and simple marker for the evaluation of UI.

Ultrasound Assessment of True Fetal Head Descent for Optimizing Operative Vaginal Delivery.

The foetal head's automatic segmentation from ultrasound imagery is considered a key step in prenatal examination. However, achieving high-quality semi-supervised foetal head image segmentation remains challenging due to low image resolution, unclear boundaries, and inconsistencies between labelled and unlabelled data. To overcome these obstacles, we propose MCPNet, a morphological constraint-based copy-paste network for semi-supervised foetal head segmentation, incorporating score-guided morphological refinement (SMR) and copy-paste mixing augmentation (CPMA). SMR employs weighted scores derived from Sobel operators and Euclidean transform to ensure boundary consistency. Additionally, to mitigate the distribution gap between labelled and unlabelled data, we introduce CPMA. This method uses random cropping to swap foreground and background between labelled and unlabelled data. On the HC18 and PSFH benchmarks, our method achieves Dice scores of 93.72% and 92.31% respectively with 20% labelled data. The results demonstrate our superior performance and clinical potential.

Adapt or specialize? A comprehensive evaluation of adapted SAM versus task-specific CNNs for fetal abdominal segmentation.

To evaluate whether averaging replicate fetal biometry measurements during an examination improves estimated fetal weight (EFW) accuracy for predicting birthweight compared with a single set and to assess how variability in fetal biometric measurements changes across gestation. Secondary analysis of the Eunice Kennedy Shriver National Institute of Child and Health and Human Development (NICHD) Fetal Growth Studies-Singletons. At each study visit, fetal head circumference (HC), abdominal circumference (AC), and femur length (FL) were measured in triplicate to compute EFW at each of 6 study visits (0-5). Variability among replicates was assessed across gestation using boxplots and two-way repeated measures ANOVA. For participants with visit 5 EFW (n = 1101), predictive accuracy of each individual measurement and their averages was compared with birthweight using average percent error and proportion within 10% of birthweight. One-way ANOVA and chi-square tests were applied. Sensitivity analyses included second-to-last visit, subgroups classified as small- or large-for-gestational age, and a subgroup of participants who delivered within 3 days of visit 5. Across visits, replicate HC, AC, and FL showed minimal variability, with no significant differences among replicates or across gestation. At visit 5, average EFW and standard deviations were similar across individual and averaged measurements. Percent errors ranged from 8.2 to 8.5% (p = .999), and proportion within 10% of birthweight ranged from 64.4 to 66.4% (p = .954). Sensitivity analyses produced similarly null findings. In a controlled research setting with credentialed sonographers and standardized protocols, averaging replicate fetal biometric measurements did not improve EFW prediction of birthweight. A single high-quality measurement of fetal biometry may be sufficient, although further research is warranted to confirm generalizability to routine clinical practice.

The measurement of fetal biometrics from ultrasound images plays a vital role in assessing potential development during pregnancy. However, existing fetal segmentation methods failed to accurately segment and asses the head circumference that gives inaccurate segmentation results. To overcome this limitation, a feature feedback and global feature with adaptive feature fusion network (FGA–Net) model is proposed to enhance fetal head segmentation (FHS). It involves four key components for feature extraction, fusion, and correction, respectively. The adaptive feature fusion module (AFFM) and correction map integrate the local features and global features and refine the features to enhance accurate FHS from the ultrasound images efficiently. Initially, ultrasound images are obtained from the two publicly available datasets and preprocessed using normalization and data augmentation techniques. Finally, preprocessed images are fed to FHS by proposed FGA Net utilizing EfficientNet-B0 as the backbone network for efficient feature extraction. Experimental results of proposed FGA-Net are evaluated using the dice coefficient (DC) of 95.78% and 98.95% for FH-PS-AoP and HC-18 datasets, which shows better results than the existing segmentation approaches like inverted bottleneck patch expanding (IBPE) method.

Vaginal deliveries are frequently associated with perineal trauma, including severe tearing in some cases. Understanding of pelvic floor muscle damage and perineal tearing during childbirth is of great clinical relevance. However, the knowledge of these complex phenomena is incomplete. The objective of the present study is to explore the multifactorial view of pelvic floor muscle damage and perineal tearing during childbirth. Using nonlinear finite element modeling coupled to statistical surrogate modeling, we modeled fetal descent with imposed displacement and used active maternal for muscle contraction to estimate the pelvic floor muscle damage and perineal tearing indicators under different influencing factors such as fetal head deformability and biometry, as well as constitutive behaviors. The obtained results show that fetal head deformability reduces stress and strain concentrations in the pelvic floor muscles (PFM) and perineal region, while increasing fetal head size leads to heightened internal tissue responses. Linear regression analysis demonstrated strong model performance (R² = 0.782-0.981) and statistically predictive relationships between fetal biometric parameters, soft tissue constitutive behaviors, and associated mechanical responses. By integrating advanced finite element modeling with statistical modeling and regression, this work provides new quantitative insights into the biomechanical factors, highlighting tissue deformation patterns and indicating potential risk of tissue damage in highly strained areas due to localized mechanical stress. This approach offers a predictive and non-invasive strategy for assessing maternal tissue vulnerability during childbirth.

B-mode ultrasound (USG) is a key imaging modality for fetal assessment, providing a noninvasive approach to monitor anatomical development and detect congenital anomalies at an early stage. However, portable ultrasound devices commonly used in low-resource healthcare settings often yield low-resolution images with significant speckle noise, reducing diagnostic accuracy. Furthermore, the scarcity of labeled medical data, caused by privacy regulations such as HIPAA and the high cost of expert annotation, poses a significant challenge for developing robust artificial intelligence (AI) diagnostic models. This study proposes a CycleGAN-based deep learning model enhanced with a histogram-guided discriminator (HisDis) to generate realistic synthetic B-mode fetal ultrasound images. A publicly available dataset from the Zenodo repository containing 1,000 grayscale fetal head images was utilized. Preprocessing included normalization, histogram equalization, and image resizing, while the architecture combined two ResNet-based generators and a dual discriminator configuration integrating PatchGAN and histogram-guided evaluation. The model was trained using standard optimization settings to ensure stable convergence. Experimental results demonstrate that the proposed HisDis module accelerates convergence by 18 epochs and reduces the Fréchet Inception Distance (FID) by 23.6 percent from 1580.72 to 1208.49 compared with the baseline CycleGAN. Statistical analysis revealed consistent pixel-intensity distributions between the original and synthetic images, with entropy from 7.16 to 7.40. At the same time, visual assessment confirmed that critical anatomical structures, including the brain midline and lateral ventricles, were well preserved. These results indicate that the CycleGAN-HisDis model produces statistically and visually realistic fetal ultrasound images suitable for medical data augmentation and AI-based diagnostic training. Furthermore, this approach holds potential to enhance diagnostic reliability and clinical education in healthcare settings with limited imaging resources. Future work will focus on clinical validation and generalization across diverse fetal ultrasound datasets.

Objectives To explore the correlation between different delivery methods for high-positioned fetal heads during cesarean section(CS) and obstetric complications. Methods A retrospective analysis of 236 cephalic-position CS patients with high-floating fetal heads (Jun 2023-Jun 2024) was conducted. Patients were divided into manual delivery (n=80), single-blade forceps (n=80), and double-blade forceps (n=76) groups. Perioperative outcomes, including delivery time, Apgar scores, and maternal/fetal complications, were compared. Results The single-blade forceps group had the shortest delivery time (p=0.000), higher 1-minute Apgar scores than the manual proup (p=0.010), and the lowest incidence of birth injury (p = 0.000). Conclusions For high-positioned fetal heads, single-blade forceps significantly shorten delivery time and reduces neonatal asphyxia, birth injury, and maternal complications.

Overcoming shoulder dystocia with the Zavanelli maneuver: a case report

Managing Impacted Fetal Head: Insights From a Simulation Study of Applied Forces.

This paper presents a case of dystocia due to Fetal Arthrogryposis Multiplex Congenita in a three-year-old Balami Ewe. The case was presented to Senator Ali Sheriff Veterinary Hospital, Maiduguri, Borno State by a client who noticed unproductive labor for approximately three (3) hours after rupture of the water bag. Physical examination of the Ewe revealed presence of fetal head and thorax in anterior longitudinal presentation with extended forelimbs hanging through the vulva, however, the hind limb was lodged in the birth canal. A pedal and suckling reflexes were observed in the lamb and obstetrical examination revealed a tail and flexed hocks impacted within the birth canal. There was adequate pelvic dimension and spacing that allowed per-vaginum manoeuvring and traction for relieve of the dystocia. A live fetus was delivered although characterized with ankylosed joints of the hind limb, deformed lumbo-sacral vertebral column, kyphosis of the thoracic spines with very poor gait when the lamb attempts to stand. In addition, the dam was weak and required support in order to stand. Supportive therapy involved intravenous infusion of 5% Dextrose saline (500mL) and intramuscular injection of Diclofenac sodium given at 2.5mg/kg body weight. Oxytocin (10I.U) was administered intramuscularly to facilitate uterine involution and reduced post-parturient hemorrhage. The fetus was placed on colostrum obtained from the dam but died within 24 hours of birth; however, the supportive therapy was continued for the dam for 3 days’ post-partum. Follow-up assessments showed that the Ewe has regained a stable condition with no signs of retained fetal membranes, septicemia or metritis. The client was advised to avoid rebreeding the ewe and to use rams with proven reproductive histories for further breeding and genetic control

Background/Aims Artificial intelligence can be used in birth monitoring, where advanced algorithms can predict outcomes, detect preterm birth risks and identify pregnancy complications. This review aimed to map the existing evidence on the use of artificial intelligence in ultrasound-based monitoring of labour progress. Methods Five databases were systematically searched: Cochrane Review Library, CINAHL, Medline, EMBASE, Scopus and Web of Science. Primary studies published between 2000 and 2025 that explored using artificial intelligence for ultrasound, vaginal examination or to estimate the progress of labour were included. Results A total of 14 articles were included, involving 145 085 women and labour/birth records. Key areas where artificial intelligence was used included automated fetal head position assessment, segmentation of anatomical structures and predicting dystocia and mode of birth. Challenges remained in validation, standardisation, regulatory approval and integration to clinical workflows. Conclusions Integrating artificial intelligence into vaginal examinations and labour monitoring could enhance safety, accuracy, and efficiency. Implications for practice Further validation with larger datasets and live patient studies is recommended before widespread clinical implementation.

Ultrasound is potentially more accurate for identifying women who are likely to have a spontaneous vaginal birth than conventional labor progression assessments. However, research on the application of intrapartum ultrasound measurements in predicting cesarean section (CS) during vaginal trials of labor in full-term low-risk nulliparous women is lacking. We aimed to develop a predictive model using intrapartum ultrasound data for CS conversion in nulliparous women who experienced a failed vaginal delivery trial and integrate this model into clinical practice. Low-risk nulliparous women carrying full-term singleton vertex fetuses were randomly divided into derivation and validation cohorts (ratio 8:2). Ultrasound was used to measure the angle of progression and head–perineum distance of a cervical dilation of 4 to 6 cm. Maternal age, gestational age, height, weight, and other factors were included. Univariate and binary logistic regression analyses were conducted to identify factors affecting failed vaginal delivery trials and subsequent CS conversion. A risk prediction model was constructed, and its discrimination, calibration, and clinical usefulness were assessed. Independent predictors of CS conversion included oxytocin use for induction, fetal head circumference, estimated fetal weight, labor analgesia, and angle of progression when the active labor phase commenced (P < .05). The nomogram showed good discrimination and calibration. The receiver operating characteristic curve area for the derivation and validation cohorts were 0.924 (95% confidence interval: 0.892–0.956) and 0.916 (95% confidence interval: 0.817–01.000), respectively, with sensitivities and specificities of 0.933 and 0.781 for the derivation cohort and 0.857 and 0.827 for the validation cohort, respectively. Concordance indexes for the derivation and validation cohorts were 0.92 and 0.91, respectively. The predictive model exhibited robust predictive capabilities and high precision. It can provide a reference for clinicians to guide women to choose the appropriate mode of delivery, thereby improving maternal and infant outcomes.

PurposeImpacted fetal head at full-dilation cesarean delivery (CD) is a major cause of adverse maternal and neonatal outcomes. The Fetal Pillow is a device designed to reduce these complications. Our study aims to evaluate the outcomes of full-dilation CD with and without the use of Fetal Pillow.MethodsThis retrospective cohort study included full-dilation CDs performed from January 2018 to July 2023, at a single tertiary center. Indications for CDs were arrest of decent, fetal distress, and failed vacuum extraction. The study cohort included cases (Fetal Pillow group) matched to controls (without the use of Fetal Pillow) according to the indication to CD, in a 1:2 ratio. The study evaluated maternal outcomes as uterine incision extensions, maternal blood loss (ml), maternal postoperative infection, and length of hospitalization (days). Neonatal outcomes included NICU admissions, cord arterial blood pH, Apgar scores, respiratory distress, intubation, and seizures.ResultsThe study included 138 patients, of them 46 in the Fetal Pillow group and 92 matched controls. Maternal baseline characteristics, including age, BMI, prior CDs, and obstetric complications, were similar between groups. No significant differences were observed in median surgery duration [55 (44–68) vs. 54 (42–65) min; p = 0.361] or fetal extraction time [6 (4–8) vs. 5.5 (3–8) min; p = 0.507]. Maternal outcomes, including rates of uterine incision extensions, blood loss, and length of hospitalization, were comparable between groups. However, NICU admissions were significantly lower in the Fetal Pillow group compared to controls (17.3% vs. 33.7%, p = 0.04). In multivariable logistic regression analysis, adjusting for parity, fetal head station and position, maternal age, prior CD, and neonatal birth weight, Fetal Pillow use was associated with reduced NICU admissions [adjusted odds ratio (aOR) 0.30; 95% CI 0.11–0.82; p = 0.019].ConclusionThe use of Fetal Pillow for impacted fetal head during full-dilation CDs may reduce NICU admissions without adversely affecting maternal outcomes.

Pre-operative ultrasound before cesarean delivery: reviewing findings during the first surgical time-out to improve maternal and infant safety.

The impacted fetal head at cesarean delivery, incidence, complications and management options, including a new device.

Is sonographic measurement of head-perineum distance useful for predicting a need for instrument change in vacuum-assisted delivery?