Three-dimensional modeling of color Doppler images: a new approach to study follicular vascularization in cattle

ABSTRACT

Ovarian blood supply is directly related to follicle developmental potential and to oocyte quality, and color Doppler ultrasonography might be a valuable tool to predict in vitro fertilization outcomes. In most studies in large domestic animals, however, the evaluation of follicle blood flow is qualitative (presence or absence of color signal) or dependent on the analysis of a single image. The objective of the present study was to first describe the use of a three-dimensional (3D) modeling of color Doppler images for a quantitative assessment of vascularization in bovine ovarian follicles. Follicular wave emergence was synchronized in Holstein and Gir heifers (n = 20), and follicular dynamics were assessed every 12 h using a color Doppler ultrasound device. The recorded cine-loop of the dominant follicle was decomposed into frames and medical image processing software was used to isolate the Doppler signal, generate the 3D model and calculate the volume of vascularization. In experiment 1, the model was validated by comparing the expected and calculated volumes and was used to predict possible variations in the results of the 2D approach. In experiment 2, vascularization was analyzed during follicular development. In both breeds, the volume of vascularization increased after follicle deviation and was positively correlated (P < 0.05) to follicular diameter (r = 0.65 and 0.54 for Holstein and Gir heifers, respectively). Spatial analysis of the three-dimensional model showed an uneven distribution of vascularization in the follicular wall, with a more intense blood flow being detected in the basal (nearest the ovarian hilus) and lateral regions of the dominant follicles. These results demonstrate the potential of this technique as a new tool for in vivo studies of ovarian physiology in large animals.