EC Gynaecology

Background: Ultrasound follicle tracking is an important part of cycle monitoring. OSIS Ovary (Online System for Image Segmentation for the Ovary) has been conceived aiming to aid the management of the workflow in follicle tracking, one of the most iterative procedures in cycle monitoring during ovarian stimulation.

Methods: In the present study, we compared OSIS Ovary (as three-dimensional ultrasound-based automated system) with the two-dimensional manual standard measurement method, in order to assess the reliability of the main measurements obtained to track follicle growth during ovarian stimulation cycles, the follicle size and count.

Results: Based on the mean follicle diameter and follicle count values obtained, the Pearson/intraclass correlation coefficients were 0.976/0.987 and 0.804/0.889 in ≥ 10 mm follicles, 0.989/0.994 and 0.809/0.867 in ≥ 13 mm follicles and 0.995/0.997 and 0.791/0.840 in ≥ 16 mm follicles. The mean difference (MnD) for the mean diameter and follicle count was, respectively, 0.759/0.161 in ≥ 10 mm follicles, 0.486/1.033 in ≥ 13 mm follicles and 0.784/0.486 in ≥ 16 mm follicles. The upper and lower limits of agreement (ULA and LLA) were 3.641/2.123 and 5.392/3.070 in ≥ 10 mm follicles, 3.496/2.522 and 4.285/2.218 in ≥ 13 mm follicles, and 3.723/2.153 and 2.432/1.459 in ≥ 16 mm follicles. The limits of agreement range (LoAR) were 5.764/8.462 in ≥ 10 mm follicles, 6.048/6.503 in ≥ 13 mm follicles and 5.876/3.891 in ≥ 16 mm follicles. P < 0.05 was considered for all calculations

Conclusion: As three-dimensional ultrasound-based automated system in comparison with two-dimensional manual method standard, we found OSIS Ovary as a reliable tool to track follicle growth during ovarian stimulation cycles.

 Keywords: Follicle; Folliculometry; Follicle Growth Tracking; Three-Dimensional Ultrasound

1. Vandekerckhove F., et al. “The value of automated follicle volume measurements in IVF/ICSI”. Frontiers in Surgery 1 (2014): 18.
2. Raine-Fenning N., et al. “SonoAVC: a novel method of automatic volume calculation”. Ultrasound in Obstetrics and Gynecology 6 (2008): 691-696.
3. Deutch TD., et al. “Automated assessment of ovarian follicles using a novel three-dimensional ultrasound software”. Fertility and Sterility 5 (2009): 1562-1568.
4. Ata B., et al. “Comparison of automated and manual follicle monitoring in an unrestricted population of 100 women undergoing controlled ovarian stimulation for IVF”. Human Reproduction 1 (2011): 127-133.
5. Raine-Fenning N., et al. “Automated follicle tracking improves measurement reliability in patients undergoing ovarian stimulation”. Reproductive BioMedicine Online 5 (2009): 658-663.
6. Ronneberger O., et al. “U-Net: Convolutional networks for biomedical image segmentation”. MICCAI2015 (2015): 234-241.
7. Milletari F., et al. “V-Net: Fully convolutional neural net-works for volumetric medical image segmentation”. Fourth International Conference on 3D Vision (3DV) (2016): 565-571.
8. Lamazou F., et al. “Reliability of automated volumetric measurement of multiple growing follicles in controlled ovarian hyperstimulation”. Fertility and Sterility 6 (2010): 2172-2176.
9. Jayaprakasan K., et al. “The interobserver reliability of off-line antral follicle counts made from stored three-dimensional ultrasound data: a comparative study of different measurement techniques”. Ultrasound in Obstetrics and Gynecology 3 (2007): 335-341.
10. Raine-Fenning N., et al. “Automated measurements of follicle diameter: a chance to standardize?” Fertility and Sterility 4 (2009a): 1469-1472.
11. Raine-Fenning N., et al. “Automated follicle tracking improves measurement reliability in patients undergoing ovarian stimulation”. Reproductive BioMedicine Online 5 (2009b): 658-663.
12. Gerris J and De Sutter P. “Self-operated endovaginal telemonitoring (SOET): a step towards more patient-centred ART?” Human Reproduction 3 (2010): 562-568.