VP62.05: Quantitative ultrasound measurement of endometrial waves in abrnormal uteri: interim results of the WAVES study

Measurement and analysis of uterine contractions (UC) through a machine-learning model based on quantitative 2D and 3D transvaginal ultrasound (TVUS) measurements in women with abnormal uteri due to adenomyosis, leiomyomas or uterine anomalies. Design: Prospective observational cohort study. Population: 11 women with adenomyosis and 9 women withnormal uteri. TVUS measurements: UC characterisation and analysis wasimplemented by a dedicated speckle-tracking algorithmwith 2D/3D TVUS; with frequency, direction, amplitudeand velocity-related features extracted from the derivedsignals. Measurements were carried out at different pointsof the menstrual cycle and in relation to hormonal therapyand IVF treatment. Significantly higher contraction frequency was found in the healthy group during the peri-ovulatory phase (p < 0.05), but not during menstruation, compared to the adenomyosis patients. Contraction amplitude was significantly higher in the adenomyosis group, suggesting stronger uterine contractility (p < 0.05). No significant differences in terms of direction or velocity were observed. A further 60 women with abnormal uteri wil be included in this study. Study outcomes so far are promising, UC can be quantitatively described using a 2D/3D TVUS machine learning model. Results in specific populations show a clinically relevant application of this tool and confirm differences in uterine movement in abnormal versus healthy uteri. This adds to the aetiological understanding of clinical symptoms of these conditions (i.e. dysmenorrhoea, infertility). Further research in women with benign uterine disorders will lead to a better understanding of abnormal uterine contractility in this population. Supporting information can be found in the online version of this abstract Supporting Information Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.