Decision Support System for human oocytes selection
In recent years, the number of couples requesting for assisted fertilization is increasing. Both scientific and legislative aspects (Italian law 40/2004) prompted researchers to focus on effective oocyte selection methods in order to choose gametes with highest quality and potential for good embryo development. In this scenario, it is of a great importance the development of a decision support system able to classify oocytes according to a score based on morphological features and patients’ clinical data that will support biologists in oocyte non-invasive selection for fertilization.
The system we have developed could:
• support effectively to solve the problem of increasing infertility
• offer to doctors and fertilization experts a more effective selection method in order to have a clear picture of all medical data and a quantitative metric able to support their experience in oocyte selection.
• help to reduce the number of cycles of ICSI required for fertilization
• meet the Italian IVF law 40 – 2004 requirements
In order to reach this goal we have:
• defined a standard image acquision protocol
• taken more than 150 oocytes images, in standard and comparable conditions, from 35 women
• defined a features extraction protocol and a standard data format
• developed an automatic morfological features extraction and organization process
• defined a scoring alghoritm and tested it on data collected
Finally, we have developed a prototype of the system that allows to upload (to the oocyte DB) images and patient's clinical data and to search/visualize them.