A semiautomated technique for evaluation of uterine peristalsis

Purpose: To evaluate the capability of a newly developed semiautomatic analysis technique for evaluation of uterine peristalsis in comparison with visual assessment. Materials and Methods: Cine magnetic resonance (MR) images were obtained from 16 normal volunteers in periovulatory phase. Sixty-eight studies were evaluated by both visual evaluation and by a semiautomated computer assisted software. The software was developed by applying an automated contour tracking method and enables objective analysis of frequency, amplitude, velocity, and direction. Assessments were based on: 1) the recognition of peristalsis, 2) peristaltic direction and frequency, and 3) the way of recognizing peristaltic direction. The data obtained by both modalities were compared with those of previously reported ultrasound (US) findings. Results: Uterine peristalsis was identified in all studies by both objective and visual analyses. There was no significant difference in the identification of peristalsis, recognition of peristaltic directions, or peristaltic frequency (P > 0.05). Data were well in accordance with those in US literature. Peristaltic direction was recognized either by configuration change in the endometrium or by conduction of low signal intensity in the subendometrial myometrium and there was statistically no difference. Conclusion: Evaluation of uterine peristalsis by the semi-automatic analysis accorded well with the results obtained visual assessments and previous US findings.