Assessing the viability of a blastosyst is still empirical and non-reproducible nowadays. We developed an algorithm based on artificial vision and machine learning (and other classifiers) that predicts pregnancy using the beta human chorionic gonadotropin (b-hCG) test from both the morphology of an embryo and the age of the patients. We employed two high-quality databases with known pregnancy outcomes (n = 221). We created a system consisting of different classifiers that is feed with novel morphometric features extracted from the digital micrographs, along with other non-morphometric data to predict pregnancy. It was evaluated using five different classifiers: probabilistic bayesian, Support Vector Machines (SVM), deep neural network, decision tree, and Random Forest (RF), using a k-fold cross validation to assess the model’s generalization capabilities. In the database A, the SVM classifier achieved an F1 score of 0.74, and AUC of 0.77. In the database B the RF classifier obtained a F1 score of 0.71, and AUC of 0.75. Our results suggest that the system is able to predict a positive pregnancy test from a single digital image, offering a novel approach with the advantages of using a small database, being highly adaptable to different laboratory settings, and easy integration into clinical practice.

To estimate the effectiveness of prophylactic and therapeutic cerclage by meta-analysis of randomized clinical trials. We searched the Cochrane Pregnancy and Childbirth Group specialized register of clinical trials (May 2002). Congress proceedings of international society meetings of fetal-maternal and reproductive medicine were searched by hand. Meta-analysis of randomized clinical trials comparing cervical cerclage with expectant management during pregnancy was performed. Further clarification was sought from trial authors when required. Six trials describing a total of 2175 women were analyzed. Prophylactic cerclage was compared with no cerclage in four trials. Pooled results failed to show a statistically significant reduction in pregnancy loss and preterm delivery rates, although a small reduction in births less than 33 weeks' gestation was seen in the largest trial (relative risk [RR] 0.75; 95% confidence interval [CI] 0.58 to 0.98). Cervical cerclage was associated with mild pyrexia, increased use of tocolytic therapy, and hospital admission but no serious morbidity. Two trials examined the role of therapeutic cerclage when ultrasound examination revealed a short cervix. Pooled results failed to show a reduction in total pregnancy loss, early pregnancy loss, or preterm delivery before 28 and 34 weeks in women assigned to cervical cerclage. The effectiveness of prophylactic cerclage in preventing preterm delivery in women at low or medium risk for second-trimester pregnancy loss has not been proven. The role of cerclage in women whose ultrasound reveals short cervix remains uncertain.

The prediction of in vitro fertilization (IVF) outcome is an imperative achievement in assisted reproduction, substantially aiding infertile couples, health systems and communities. To date, the assessment of infertile couples depends on medical/reproductive history, biochemical indications and investigations of the reproductive tract, along with data obtained from previous IVF cycles, if any. Our project aims to develop a novel tool, integrating omics and artificial intelligence, to propose optimal treatment options and enhance treatment success rates. For this purpose, we will proceed with the following: (1) recording subfertile couples’ lifestyle and demographic parameters and previous IVF cycle characteristics; (2) measurement and evaluation of metabolomics, transcriptomics and biomarkers, and deep machine learning assessment of the oocyte, sperm and embryo; (3) creation of artificial neural network models to increase objectivity and accuracy in comparison to traditional techniques for the improvement of the success rates of IVF cycles following an IVF failure. Therefore, “omics” data are a valuable parameter for embryo selection optimization and promoting personalized IVF treatment. “Omics” combined with predictive models will substantially promote health management individualization; contribute to the successful treatment of infertile couples, particularly those with unexplained infertility or repeated implantation failures; and reduce multiple gestation rates.

We wanted to test the hypothesis that using abdominal ultrasound at the time of embryo transfer to guide replacement, improved pregnancy rates by at least 5%. An RCT in a large assisted conception unit. A pilot study and power calculation suggested that at least 2000 embryo transfers were required to demonstrate a difference of 5%, for a test with 80% power and Type 1 error 0.05. Randomization, data entry and analysis were arranged independently. Randomization was stratified for age and fresh/frozen embryo transfer. Analysis was by intention to treat. There was no difference in clinical pregnancy or live birth rates between the two groups. The clinical pregnancy rate for ultrasound-guided embryo transfer was 22% and for non-ultrasound-guided embryo transfer was 23% (odds ratio: 0.96; 95% confidence interval: 0.79–1.18). We set out to determine whether ultrasound-guided embryo transfer improved clinical pregnancy rates and live birth rates in assisted conception. We used an appropriately powered RCT design. We did not demonstrate a difference. This outcome is at odds with the UKs National Institute of Clinical Excellence recommendations for fertility treatment (Fertility Assessment and Treatment for People with Fertility Problems. London, UK: RCOG Press, 2004, 112.) which used a meta-analysis of four smaller trials (range 362–800 patients, totalling 2051 embryo transfers) to conclude that ultrasound should be offered. We suggest that the current Cochrane review should be updated with data from our trial and recommend that consideration is given to accounting for heterogeneity between the included trials.