[home] [Personal Program] [Help]

---

Session: Poster session II
Session starts: Thursday 24 January, 16:00

---

Tom Bakkes ()
Federica Sammali ()
Nienke Kuijsters ()
Chiara Rabotti ()
Dick Schoot ()
Massimo Mischi ()

Abstract:
The increasing trend towards postponing childbirth has led to an increase in infertility risk [1]. Currently, around 1 in 5 couples have problems conceiving or maintaining an established pregnancy [2]. The hope of these couples often relies on assisted reproductive technologies, such as in-vitro fertilization (IVF). However, IVF failure rate is still above 70%. Although the causes for IVF failure are still not well understood, uterine receptivity seems a determinant factor for successful conception. Understanding the role of contractions outside pregnancy and possibly quantifying them may help understanding the link between normal uterine contractility and infertility, especially in relation to IVF failure. Recently, electrohysterography (EHG) and ultrasound (US) speckle tracking have shown to be promising for non-invasive, quantitative assessment of uterine activity. In this study, we investigated the use of machine learning for discriminating the uterine activity during the four phases of the menstrual cycle (menses, late follicle, early luteal, and late luteal phase). EHG and US acquisitions were performed on 6 healthy women in 4 phases of the menstrual cycle (total of 24 observations). A set of amplitude- and frequency-features were extracted from US and EHG data [3]. Four different types of classifiers were tested: support vector machine (SVM), K-nearest neighbours, Gaussian mixture model, and naïve Bayes. A full search was used to find the optimal combination of features and classifier parameters. Validation was performed by the leave-one-out method. According to the obtained results, the SVM classifier showed the best performance, resulting in an accuracy, sensitivity and specificity of 90%, 79% and 93%, respectively, by combining 2 US and 1 EHG feature. [1] A. A. de Graaff et al., Fertility and sterility( 95), 2011. [2] G. F. Whitman-Elia et al., J Am Board Fam Pract(14), 2001. [3] F. Sammali et al., Reprod Sci, 2018. [4] F. Sammali et al., IEEE Trans Ultrason Ferroelectr Freq Control, 2018