English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In-vitro fertilization (IVF) is the most advanced treatment for infertility problems; however, its failure rate is still above 70% and the exact causes are often unknown. There is increasing evidence of the involvement of uterine contractions in IVF failure, especially during and after embryo transfer (ET). In this paper, we propose a new method to predict the success of IVF based on quantitative features extracted from electrohysterography (EHG) and B-mode transvaginal ultrasound (TVUS) recordings. To this end, probabilistic classification of the uterine activity, as either favorable or adverse to embryo implantation, is investigated using machine learning. Prior to machine learning, an additional method for EHG and TVUS feature extraction is here proposed that is based on singular value decomposition of the acquired EHG and TVUS recordings. Sixteen women were measured during three phases of the IVF treatment: follicular stimulation (FS), one hour before embryo transfer (ET1), and five to seven days after ET (ET5-7). After feature space reduction by correlation filtering, three machine-learning models, namely, support vector machine (SVM), K-nearest neighbors (KNN), and Gaussian mixture model (GMM), were optimized and tested by nested leave-one-out cross validation for their ability to predict successful embryo implantation. The highest accuracy (93.8%) was achieved by KNN in all phases and by SVM and in the FS and ET1 phases. Contraction frequency, unnormalized first moment and standard deviation, obtained from EHG and TVUS analysis, were the best features selected by the three classifiers. Our results show a multi-modal, multi-parametric strategy based on quantitative features to represent a novel, promising option for prediction of successful embryo implantation, overcoming the limitations of alternative approaches based on qualitative assessment of clinical variables. Yet, a larger dataset is required for improved training of the classifiers, as well as to assess their clinical value in the context of IVF procedures.

Multi-Modal Uterine-Activity Measurements for Prediction of Embryo Implantation by Machine Learning