The improvement of oocyte selection for social freezing application

Lifestyle, educational opportunities, career choices and new unions lead to pregnancy in more advanced age, increasing the emerging preventive solution to freeze oocytes at a young age for later use. In this scenario, the oocyte selection has a great importance in order to choose the best ones capable of a good subsequent embryo development and implantation. The aim of this study was to develop a decision support system, able to classify oocytes according to a score based on morphological features and patients' clinical data. The approach would offer a more effective selection method because it is not dependent on the doctor's experience or on an "at-first-sight" impression. As a first step, a prototype system able to support embryologists in oocyte selection was presented and an experimental evaluation on a real set of data provided. The developed pipeline included the identification of main morphological features influencing oocyte quality and the assignment of a weight and of a better way of measuring them. After that, a standard data format collecting in an organized way all morphological features of oocytes, zigote and embryos and patients' clinical data was developed. More than 150 oocytes images, taken in standard and comparable conditions, from 35 women were collected. Morphological features were extracted manually and automatically. A preliminary version of the scoring algorithm was tested on these data.