Artificial intelligence in prenatal diagnosis: Down syndrome risk assessment with the power of gradient boosting-based machine learning algorithms - Turkish Journal of Obstetrics and Gynecology

Purpose: The aim of this study is to evaluate the neurodevelopmental prognosis of fetuses with congenital heart defects by considering aortic flow and oxygen saturation in the ascending aorta, and to investigate them in terms of cortical developmental delay.Materials and Methods: In this retrospective study, neurosonographic and hemodynamic parameters were measured in fetuses diagnosed with congenital heart defects. These measurements were compared with those of gestational-age-matched low-risk controls in the second and third trimesters. Standard fetal biometry, transcerebellar diameter, Sylvian fissure area, and related ratios were evaluated.Results: The study included 53 fetuses with congenital heart defects and 72 controls in the second trimester, and 72 fetuses with congenital heart defects and 58 controls in the third trimester. In the second trimester, the Sylvian fissure area was lower in the CHD group (50.98 ± 21.02 mm² vs 58.37 ± 16.41 mm²), while the HC/SYL ratio was higher in the CHD group in both the second trimester (4.53 ± 1.69 vs 3.44 ± 0.74) and the third trimester (2.71 ± 1.31 vs 2.09 ± 0.49). Neurodevelopmental parameters showed that the Sylvian fissure area was smaller in the second-trimester congenital heart defects group compared with controls. Consequently, the aortic valve diameter/Sylvian fissure area and head circumference/Sylvian fissure area ratios were higher in the congenital heart defects group. In the third trimester, patients with normal aortic flow and two functional ventricles had higher aortic valve diameter/Sylvian fissure area ratios than other patient groups. Conclusion: Detecting early-onset cortical developmental delay in patients with left cardiac dysfunction using proportional data is crucial for a holistic approach to evaluation. Incorporating head circumference/Sylvian fissure area and aortic valve diameter/Sylvian fissure area ratios into routine prenatal follow-up measurements may help identify fetuses at potential risk for adverse postnatal neurodevelopmental outcomes.

Background/Aim: This study aims to develop an interpretable and practical decision support method for early prediction of the need for cesarean delivery. Although machine learning and deep learning models are prevalent in the literature, their generalization capabilities are often restricted, especially when utilizing small clinical datasets. This limitation underscores the necessity for robust, transparent, and well-regularized models in medical decision-making processes. Methods: The study proposed a stacking-based hybrid model, which combines the strengths of both classical and modern techniques. The data were normalized using StandardScaler, and feature selection involved principal component analysis (PCA) and SelectKBest to capture global and target-relevant patterns. In the classification phase, two parallel learners – a regularized multi-layer perceptron (MLP) and logistic regression – were used, followed by a random forest meta-learner. Results: The experimental analysis demonstrated that the proposed model achieved an average accuracy of 96.43% under stratified 5-fold cross-validation. Although this result surpassed the performance of other baseline models within the dataset, it should be regarded as preliminary due to the limited sample size. Conclusion: The findings indicate that the proposed hybrid approach has potential as a promising direction for future clinical decision support research. Nonetheless, additional validation using larger and more diverse datasets is necessary to adequately assess its generalizability and practical utility.

Background and Objectives: To evaluate pregnancy outcomes and transthoracic echocardiographic (TTE) findings during the antenatal and postpartum periods in women with repaired Tetralogy of Fallot (ToF) who delivered at Çukurova University Faculty of Medicine, Balcalı Hospital, between 2011 and 2025 and to interpret these findings in the context of the 2025 European Society of Cardiology (ESC) recommendations. Materials and Methods: This single-center retrospective cohort study undertaken between 2011 and 2025 included 11 pregnant women with surgically repaired ToF. Maternal demographic characteristics, obstetric outcomes, mode of delivery, neonatal outcomes, and antenatal TTE parameters were recorded. Cardiac measurements obtained during pregnancy were compared with postpartum TTE findings performed 3–6 months after delivery to assess pregnancy-related cardiac adaptation and recovery. Results: A total of 11 pregnancies in women with repaired ToF were analyzed. Nine pregnancies resulted in live births, while one ended in missed abortion at 9 + 2 weeks and one in intrauterine fetal demise at 34 + 2 weeks. Among live births, the mean gestational age was 36 + 2 weeks and the mean birthweight was 2865 g, with a preterm delivery rate of 55.6%. Cesarean section was performed in 70% of cases, while 30% delivered vaginally. During pregnancy, the mean left ventricular ejection fraction was 62.6%, and residual tricuspid regurgitation was the most frequently observed echocardiographic finding. Postpartum TTE evaluations indicated that echocardiographic parameters were largely stable over the observation period, with no numerical change and no clear evidence of deterioration in ventricular function or progression of valvular regurgitation. Conclusions: Despite successful surgical repair, pregnancy may still pose potential risks for women with ToF, underscoring the importance of individualized, multidisciplinary management. In this cohort, pregnancy appeared to be generally well-tolerated when care was provided in accordance with contemporary ESC recommendations. The observation of preserved ejection fraction and overall stable right ventricular function in the early postpartum period suggests that favorable maternal cardiac adaptation may be achievable in carefully selected patients. Early postpartum echocardiographic assessment may be useful for identifying functional changes and informing structured long-term follow-up strategies.

Clinical and mycological data are essential for the optimal management of patients with Candida vaginitis (CV), particularly in cases of (i) azole-resistant C. albicans vaginitis, (ii) recurrent CV, and (iii) CV in pregnant women. The present retrospective single-center study investigated the antifungal activity of six commonly used antifungals against randomly selected vaginal isolates recovered from 68 pregnant women in Adana, Türkiye, including C. albicans, petite C. glabrata, non-petite C. glabrata, and C. krusei, using the disk diffusion method at pH 4 and 7. Furthermore, the antifungal activities of fluconazole and itraconazole were also assessed using the broth microdilution method. For all isolates, the mean inhibition zone diameters were narrower for itraconazole and ketoconazole and larger for miconazole at pH 4 than pH 7 (p < 0.05). For nystatin, zone diameters were wider in C. albicans and petite C. glabrata at pH 4 (p < 0.001 and p < 0.001). Remarkably, clotrimazole was more active at pH 4 than at pH 7, except against non-petite C. glabrata isolates. Based on the broth microdilution results, the resistance rate was higher at pH 4 than at pH 7 in all isolates. Candida glabrata petite isolates exhibited MIC values 2 to 5 times higher than those of the non-petite isolates for both fluconazole and itraconazole. This study highlights the potent activity of topical antifungals (miconazole, nystatin, and clotrimazole) for the treatment of CV in pregnant women and highlights the need to identify petite and non-petite mutants of vaginal C. glabrata isolates to obtain more reliable data and for antifungal susceptibility testing prior to decision-making. The results of the two antifungal susceptibility methods were compared for C. albicans and C. glabrata isolates, and the reliability of the disk diffusion test was discussed.