Automatic semantic segmentation of EHG recordings by deep learning: An approach to a screening tool for use in clinical practice.

BACKGROUND AND OBJECTIVE: Preterm delivery is an important factor in the disease burden of the newborn and infants worldwide. Electrohysterography (EHG) has become a promising technique for predicting this condition, thanks to its high degree of sensitivity. Despite the technological progress made in predicting preterm labor, its use in clinical practice is still limited, one of the main barriers being the lack of tools for automatic signal processing without expert supervision, i.e. automatic screening of motion and respiratory artifacts in EHG records. Our main objective was thus to design and validate an automatic system of segmenting and screening the physiological segments of uterine origin in EHG records for robust characterization of uterine myoelectric activity, predicting preterm labor and help to promote the transferability of the EHG technique to clinical practice. METHODS: For this, we combined 300 EHG recordings from the TPEHG DS database and 69 EHG recordings from our own database (Ci2B-La Fe) of women with singleton gestations. This dataset was used to train and evaluate U-Net, U-Net++, and U-Net 3+ for semantic segmentation of the physiological and artifacted segments of EHG signals. The model's predictions were then fine-tuned by post-processing. RESULTS: U-Net 3+ outperformed the other models, achieving an area under the ROC curve of 91.4 % and an average precision of 96.4 % in detecting physiological activity. Thresholds from 0.6 to 0.8 achieved precision from 93.7 % to 97.4 % and specificity from 81.7 % to 94.5 %, detecting high-quality physiological segments while maintaining a trade-off between recall and specificity. Post-processing improved the model's adaptability by fine-tuning both the physiological and corrupted segments, ensuring accurate artifact detection while maintaining physiological segment integrity in EHG signals. CONCLUSIONS: As automatic segmentation proved to be as effective as double-blind manual segmentation in predicting preterm labor, this automatic segmentation tool fills a crucial gap in the existing preterm delivery prediction system workflow by eliminating the need for double-blind segmentation by experts and facilitates the practical clinical use of EHG. This work potentially contributes to the early detection of authentic preterm labor women and will allow clinicians to design individual patient strategies for maternal health surveillance systems and predict adverse pregnancy outcomes.

Combination of Feature Selection and Resampling Methods to Predict Preterm Birth Based on Electrohysterographic Signals from Imbalance Data.

Due to its high sensitivity, electrohysterography (EHG) has emerged as an alternative technique for predicting preterm labor. The main obstacle in designing preterm labor prediction models is the inherent preterm/term imbalance ratio, which can give rise to relatively low performance. Numerous studies obtained promising preterm labor prediction results using the synthetic minority oversampling technique. However, these studies generally overestimate mathematical models' real generalization capacity by generating synthetic data before splitting the dataset, leaking information between the training and testing partitions and thus reducing the complexity of the classification task. In this work, we analyzed the effect of combining feature selection and resampling methods to overcome the class imbalance problem for predicting preterm labor by EHG. We assessed undersampling, oversampling, and hybrid methods applied to the training and validation dataset during feature selection by genetic algorithm, and analyzed the resampling effect on training data after obtaining the optimized feature subset. The best strategy consisted of undersampling the majority class of the validation dataset to 1:1 during feature selection, without subsequent resampling of the training data, achieving an AUC of 94.5 ± 4.6%, average precision of 84.5 ± 11.7%, maximum F1-score of 79.6 ± 13.8%, and recall of 89.8 ± 12.1%. Our results outperformed the techniques currently used in clinical practice, suggesting the EHG could be used to predict preterm labor in clinics.

Diagnosis of intraamniotic inflammation by measuring vaginal interleukin-6 in patients with cervical insufficiency: could amniocentesis be avoided?

BACKGROUND: Cervical insufficiency is a recurrent, passive, and painless dilation of the cervix in the second trimester. The etiology is unclear, but there may be an association with subclinical intraamniotic infection. Interleukin-6 (IL-6) production in the amniotic cavity is induced by bacterial invasion, it is the major proinflammatory cytokine released in response to infection. Although the gold standard method to measure it is through an amniocentesis, the procedure constitutes an invasive technique with several associated risks. The objective of this study is to determine if there is a correlation between intraamniotic and vaginal IL-6 in patients with cervical insufficiency and bulging membranes during the second trimester of pregnancy, in order to avoid an amniocentesis before the rescue cerclage. METHODS: A cohort study was performed in which all patients with cervical insufficiency and bulging membranes admitted into our tertiary hospital between 2019 and 2020 were included, and a control group of asymptomatic women in the second trimester of gestation where studied at the same time. Patients with bulging membranes underwent an amniocentesis to quantify amniotic IL-6, and a sample of vaginal fluid for vaginal IL-6 determination was obtained from both the study and the control group. RESULTS: A total of 20 women were included in each group. Median gestational age at diagnosis was 22 weeks in patients with bulging membranes, and 21 weeks in the control group. Vaginal IL-6 in control group (10.875 pg/mL) is much lower than the study group one (1308.77 pg/ml). In patients with bulging membranes, vaginal IL-6 expression was lower in the vagina than in the amniotic cavity [average IL-6 in the amniotic cavity 26890.07 pg/mL, vs 1308.77 pg/mL in the vagina (p < .01)]. Through a Spearman coefficient correlation rank [rho = 0.709 (p < .001)], there is a positive correlation between amniotic and vaginal IL-6 values. The best value of this correlation was calculated with the ROC curve, being the area under the curve 0.929 (CI 95% 0.721-0.995), and the cutoff of point less than 61.4 pg/ml (sensitivity 83.33%; specificity 92.86%). Patients with vaginal IL-6 < 61.4 pg/ml associated a longer latency time between diagnosis and delivery, a higher neonatal weight and a lower perinatal mortality. Rescue cerclage in vaginal IL-6 < 61.4 pg/ml was the best predictor of good pregnancy outcome. CONCLUSION: There is a correlation between intraamniotic and vaginal IL-6 in patients with cervical insufficiency and bulging membranes during the second trimester of pregnancy. However, further studies are needed in order to considerate the avoidance of an amniocentesis before an emergency cerclage.

Assessment of Dispersion and Bubble Entropy Measures for Enhancing Preterm Birth Prediction Based on Electrohysterographic Signals.

One of the remaining challenges for the scientific-technical community is predicting preterm births, for which electrohysterography (EHG) has emerged as a highly sensitive prediction technique. Sample and fuzzy entropy have been used to characterize EHG signals, although they require optimizing many internal parameters. Both bubble entropy, which only requires one internal parameter, and dispersion entropy, which can detect any changes in frequency and amplitude, have been proposed to characterize biomedical signals. In this work, we attempted to determine the clinical value of these entropy measures for predicting preterm birth by analyzing their discriminatory capacity as an individual feature and their complementarity to other EHG characteristics by developing six prediction models using obstetrical data, linear and non-linear EHG features, and linear discriminant analysis using a genetic algorithm to select the features. Both dispersion and bubble entropy better discriminated between the preterm and term groups than sample, spectral, and fuzzy entropy. Entropy metrics provided complementary information to linear features, and indeed, the improvement in model performance by including other non-linear features was negligible. The best model performance obtained an F1-score of 90.1 ± 2% for testing the dataset. This model can easily be adapted to real-time applications, thereby contributing to the transferability of the EHG technique to clinical practice.

Optimized Feature Subset Selection Using Genetic Algorithm for Preterm Labor Prediction Based on Electrohysterography.

Electrohysterography (EHG) has emerged as an alternative technique to predict preterm labor, which still remains a challenge for the scientific-technical community. Based on EHG parameters, complex classification algorithms involving non-linear transformation of the input features, which clinicians found difficult to interpret, were generally used to predict preterm labor. We proposed to use genetic algorithm to identify the optimum feature subset to predict preterm labor using simple classification algorithms. A total of 203 parameters from 326 multichannel EHG recordings and obstetric data were used as input features. We designed and validated 3 base classifiers based on k-nearest neighbors, linear discriminant analysis and logistic regression, achieving F1-score of 84.63 ± 2.76%, 89.34 ± 3.5% and 86.87 ± 4.53%, respectively, for incoming new data. The results reveal that temporal, spectral and non-linear EHG parameters computed in different bandwidths from multichannel recordings provide complementary information on preterm labor prediction. We also developed an ensemble classifier that not only outperformed base classifiers but also reduced their variability, achieving an F1-score of 92.04 ± 2.97%, which is comparable with those obtained using complex classifiers. Our results suggest the feasibility of developing a preterm labor prediction system with high generalization capacity using simple easy-to-interpret classification algorithms to assist in transferring the EHG technique to clinical practice.

Transitory Fetal Skin Edema in a Pregnant Patient with a Mild SARS-CoV-2 Infection.

BACKGROUND: Vertical transmission of the Coronavirus Disease 2019 (COVID-19) is still controversial. Additionally, the consequences of the infection during pregnancy in the offspring also are unknown. CASE: A transitory fetal skin edema and polyhydramnios have been demonstrated by ultrasound in a pregnant patient with COVID-19 after a negative RT-PCR for SARS-CoV-2. The fetal findings presented a spontaneous resolution in utero, and abnormal findings were not found in the newborn. CONCLUSION: Women who have undergone SARS-CoV-2 infection during pregnancy should receive a subsequent appropriate follow-up in order to clarify the fetal consequences of the novel coronavirus, if any.

Risk factors for vertical transmission of hepatitis C virus: a single center experience with 710 HCV-infected mothers.

OBJECTIVE: The aim of this study was to analyze the risk factors on the perinatal transmission of hepatitis C virus (HCV). STUDY DESIGN: A retrospective cohort study with 711 infants born to 710 HCV-infected mothers was conducted at the Hospital La Fe, in Valencia, Spain, from 1986 to 2011. As potential risk factors for transmission we analyzed: maternal age, mode of acquisition of HCV infection, HIV co-infection, antiretroviral treatment against HIV, CD4 cell count, HIV and HCV viral load, liver enzyme levels during pregnancy, smoking habit, gestational age, intrapartum invasive procedures, length of rupture of membranes, length of labor, mode of delivery, episiotomy, birth weight, newborn gender and type of feeding. RESULTS: Overall perinatal HCV transmission rate was 2.4%. The significant risk factors related with HCV transmission were maternal virus load >615copies/mL (OR 9.3 [95% CI 1.11-78.72]), intrapartum invasive procedures (OR 10.1 [95% CI 2.6-39.02]) and episiotomy (OR 4.2 [95% CI 1.2-14.16]). HIV co-infection and newborn female were near significance (p=0.081 and 0.075, respectively). CONCLUSIONS: Invasive procedures as fetal scalp blood sampling or internal electrode and episiotomy increase vertical transmission of HCV, especially in patients with positive HCV RNA virus load at delivery.

Good prognosis of cerclage in cases of cervical insufficiency when intra-amniotic inflammation/infection is ruled out.

OBJECTIVE: To determine if absence of sub-clinical intra-amniotic inflammation improves the prognosis of rescue cerclage in cases of bulging membranes. METHODS: Cohort study with all women with bulging membranes admitted into our hospital between 2009 and 2013. Patients underwent amniocentesis to quantify amniotic glucose, leukocytes, IL-6 and leukocyte esterase levels and for microbiological culture. All patients without intra-amniotic inflammation or sub-clinical chorioamnionitis were proposed a physical examination-indicated cervical cerclage. Those who did not accept were treated with bed rest. RESULTS: We enrolled 31 women. Median gestational age at diagnosis was 23 + 1 (21­25 + 4) weeks. Median interval until delivery was 12 (3­52.5) d. IL-6 had the highest diagnostic accuracy for good prognosis. Patients with IL6 <2.90 ng/ml were diagnosed later in pregnancy and presented a longer interval until delivery (89 versus 4 d), higher gestational age at delivery (35 + 1 versus 23 + 3 weeks) and a lower rate of prematurity (54.5% versus 100%) and perinatal mortality (0% versus 80%) than those with IL-6 ≥2.90 ng/ml. Rescue cerclage and low Il-6 were the best predictors of good outcome. CONCLUSION: IL-6 levels in amniotic fluid may be of clinical value for individualizing the management of patients with bulging membranes for placement of rescue cerclage.

Arabin cerclage pessary as a treatment of an acute urinary retention in a pregnant woman with uterine prolapse.

A 35-year-old gravida 7, para 1, and abortus 5 female with hypogastric pain and inability to void urine after 14 + 3 weeks of amenorrhea was examined in the emergency department. One year before, a uterine prolapse had been diagnosed in another hospital. Examination showed a uterine prolapse grade 2 with palpable bladder. The patient was unable to void urine. After a manual reduction of the uterine prolapse, the patient underwent an emergency catheterization for bladder drainage. A Hodge pessary (size 70) was placed, which led to spontaneous micturitions. Due to the persistence of the symptoms the following day, Hodge pessary was replaced by an Arabin cerclage pessary. Although the pessary could be removed from the beginning of the second trimester, due to the uterine prolapse as a predisposing factor in the patient and the uncomplicated progression of pregnancy, it was decided to maintain it in our patient. Therefore, Arabin cerclage pessary allowed a successful pregnancy outcome and was not associated with threatened preterm delivery or vaginal infection.

Pregnancy outcomes after kidney transplantation-immunosuppressive therapy comparison.

OBJECTIVE: To assess maternal, neonatal and graft outcomes after pregnancy in patients with kidney transplantation, and to compare the immunosuppressive therapies used. METHODS: Review of 29 pregnancies in 23 patients with kidney transplantation, managed at La Fe University Hospital, Valencia. Immunosuppressive therapies with Cyclosporine-A, Tacrolimus, Mycophenolate mofetil and Azathioprine were compared. RESULTS: No statistical differences were found in perinatal or maternal complications, with respect to the immunosuppressive therapy used. There were no differences between therapy and graft survival. Maternal complications occurred in 25 out of 28 deliveries. The most common were anemia (75%) and hypertension (53.6%). Of the 29 pregnancies, 26 were live deliveries, two were stillbirths and one was a miscarriage. The median birth weight of newborns was 2650 g (900-4350 g). From the 28 deliveries, maternal complications were reported in 25 patients. Perinatal complications were recorded in 55.6% of the patients, with prematurity being the most common (44.4%) type. One malformation was reported, this was a cleft palate in a 25 year old patient who was treated with mycophenolate mofetil. CONCLUSION: Pregnancies in patients with kidney transplantation should be considered high-risk pregnancies because of the higher rate of maternal and perinatal complications. Immunosuppressive therapies have not shown differences in maternal or perinatal outcomes.