Artificial intelligence (AI) has gained prominence in medical imaging, particularly in obstetrics and gynecology (OB/GYN), where ultrasound (US) is the preferred method. It is considered cost effective and easily accessible but is time consuming and hindered by the need for specialized training. To overcome these limitations, AI models have been proposed for automated plane acquisition, anatomical measurements, and pathology detection. This study aims to overview recent literature on AI applications in OB/GYN US imaging, highlighting their benefits and limitations. For the methodology, a systematic literature search was performed in the PubMed and Cochrane Library databases. Matching abstracts were screened based on the PICOS (Participants, Intervention or Exposure, Comparison, Outcome, Study type) scheme. Articles with full text copies were distributed to the sections of OB/GYN and their research topics. As a result, this review includes 189 articles published from 1994 to 2023. Among these, 148 focus on obstetrics and 41 on gynecology. AI-assisted US applications span fetal biometry, echocardiography, or neurosonography, as well as the identification of adnexal and breast masses, and assessment of the endometrium and pelvic floor. To conclude, the applications for AI-assisted US in OB/GYN are abundant, especially in the subspecialty of obstetrics. However, while most studies focus on common application fields such as fetal biometry, this review outlines emerging and still experimental fields to promote further research.

Purpose: Evaluating procedure-related complications and perinatal outcomes after intrauterine transfusion (IUT) before or after 20⁺⁰ weeks of gestation in fetuses with severe anemia due to intrauterine human parvovirus B19 infection. Methods: A retrospective study investigating fetuses requiring IUT for fetal Parvo B19 infection in two tertiary referral centers between December 2002 and December 2021. Procedure-related complications, intrauterine fetal death (IUFD), and perinatal outcome were correlated to gestational age (GA) at first IUT, the presence of hydrops and fetal blood sampling results. Results: A total of 186 IUTs were performed in 103 fetuses. The median GA at first IUT was 19⁺³ (13⁺⁰–31⁺⁴) weeks of gestation. IUFD occurred in 16/103 fetuses (15.5%). Overall survival was 84.5% (87/103). Hydrops (p = 0.001), lower mean hemoglobin at first IUT (p = 0.001) and low platelets (p = 0.002) were strongly associated with IUFD. There was no difference observed in fetuses transfused before or after 20⁺⁰ weeks of gestation. Conclusion: IUT is a successful treatment option in fetuses affected by severe anemia due to parvovirus B19 infection in specialized centers. In experienced hands, IUT before 20 weeks is not related to worse perinatal outcome.

Objectives: Furcate cord insertion is a rare abnormality affecting approximately 0.1% of all pregnancies. Macroscopically, the umbilical vessels separate before reaching the placenta, lose their Wharton’s jelly, and insert at the placenta centrally, eccentrically, or marginally. The aim of this retrospective study was to determine the prevalence of furcate cord insertion more accurately, the pathological characteristics, and clinical outcomes. Methods: We conducted a retrospective study of 132 cases of furcate insertion of the umbilical cord using the pathological database of the Charité University Hospital Berlin, Germany, between 1993 and 2016. This included 99 cases, including one termination of pregnancy within our institution and 33 cases from external hospitals. An analysis of the pathological features of the 132 cases and the perinatal outcome of the 98 cases within our institution were performed. Results: Furcate cord insertion occurred in 0.16% pregnancies. Of the 132 cases, seven cases of intrauterine fetal deaths were observed. Three of those could be linked to the furcate cord insertion. In two of those cases, single umbilical vessel rupture was identified as the cause of fetal death. Conclusions: In most cases of furcate cord insertion, the outcome is good; however, intrauterine fetal death occurs in approximately 1.02% of cases.

Background: Cordocentesis is used in clinical situations in which lower-risk diagnostic procedures do not deliver the desired results. The aim of this study was to evaluate the risk for procedure-related complications and fetal loss in correlation to maternal risk factors. Methods: This is a multicenter retrospective study investigating the complications, risk factors and perinatal outcome of diagnostic cordocentesis between 1998 and 2019 in three different centers. Results: A total of 1806 cordocenteses were performed and procedure-related complications (IUFD within 48 h, contractions, bradycardia, unsuccessful puncture, chorioamniotic separation) were noted in 1.6% of cases. Fetuses with chromosomal aberrations, intrauterine growth restriction and hydropic fetuses had a significantly higher rate of fetal loss compared to other indications. Fetal blood sampling (FBS) performed before 17⁺⁰ weeks of gestation was associated with a higher risk of procedure-related complications. Maternal BMI ≥ 40 increased the risk for fetal loss, whereas maternal age, number of previous miscarriages, number of previous abortions, history of vaginal bleeding or nicotine abuse did not affect the risk for complications or overall fetal loss rate. Conclusions: In the hands of experienced operators, FBS is a safe way to further fetal diagnostics, and the risk of complications is low.

Background: To report on prophylactic therapy for hyperfibrinolysis with tranexamic acid (TXA) during expectant management (EM) in the placenta accreta spectrum (PAS). Methods: This is a monocentric retrospective study of women with PAS presenting at our hospital between 2005 and 2021. All data were retrospectively collected through the departmental database. Results: 35 patients with PAS were included. EM was planned in 25 patients prior to delivery. Complete absorption of the retained placenta was seen in two patients (8%). Curettage was performed in 14 patients (56%). A hysterectomy (HE) was needed in seven (28%) patients; 18 patients (72%) underwent uterus-preserving treatment without severe complications. The mean duration of EM was 107 days. The mean day of onset of hyperfibrinolysis and beginning of TXA treatment was day 45. The mean nadir of fibrinogen level before TXA was 242.4 mg/dL, with a mean drop of 29.7% in fibrinogen level. Conclusions: Our data support EM as a safe treatment option in PAS. Hyperfibrinolysis can be a cause of hemorrhage during EM and can be treated with TXA. To our knowledge, this is the first cohort of patients with EM of PAS in whom coagulation monitoring and use of TXA have been shown to successfully treat hyperfibrinolysis.

Background: Portal hypertension in pregnancy is characterized by an increased perinatal and maternal complication rate. The purpose of this study was to evaluate the perinatal and maternal outcomes of these high-risk pregnancies at our tertiary center. Methods: We identified pregnancies with portal hypertension in our departmental database for the years 2013 to 2021. The medical history and perinatal and maternal data were extracted from medical records. Results: Eleven cases were identified. In pregnancy, delivery and postpartum, complications occurred in 72.7% of cases and included among others ascites, subclavian thrombosis, variceal-ligation-induced ulcer bleeding and postoperative hemorrhage. The cesarean delivery rate was 72.7% (n = 8); five of these were done for obstetric or fetal indications. The rate of preterm birth and admissions to neonatal intensive care unit were high (54.5% and 45.5%, respectively). Conclusions: Our case series substantiates the high maternal and perinatal complication rates seen in portal hypertension. The prevention of thromboembolic and bleeding complications was the main challenge. Care by an interdisciplinary team of experts is crucial for a successful perinatal and maternal outcome.