Uncertainty in clinical data and stochastic model for in vitro fertilization.

Abstract

In vitro fertilization (IVF) is the most widely used technique in assisted reproductive technologies (ART). It has been divided into four stages; (i) superovulation, (ii) egg retrieval, (iii) insemination/fertilization and (iv) embryo transfer. The first stage of superovulation is a drug induced method to enable multiple ovulation, i.e., multiple follicle growth to oocytes or matured follicles in a single menstrual cycle. IVF being a medical procedure that aims at manipulating the biological functions in the human body is subjected to inherent sources of uncertainty and variability. Also, the interplay of hormones with the natural functioning of the ovaries to stimulate multiple ovulation as against single ovulation in a normal menstrual cycle makes the procedure dependent on several factors like the patient's condition in terms of cause of infertility, actual ovarian function, responsiveness to the medication, etc. The treatment requires continuous monitoring and testing and this can give rise to errors in observations and reports. These uncertainties are present in the form of measurement noise in the clinical data. Thus, it becomes essential to look at the process noise and account for it to build better representative models for follicle growth. The purpose of this work is to come up with a robust model which can project the superovulation cycle outcome based on the hormonal doses and patient response in a better way in presence of uncertainty. The stochastic model results in better projection of the cycle outcomes for the patients where the deterministic model has some deviations from the clinical observations and the growth term value is not within the range of '0.3-0.6'. It was found that the prediction accuracy was enhanced by more than 70% for two patients by using the stochastic model projections. Also, in patients where the prediction accuracy did not increase significantly, a better match with the trend of the clinical data was observed in case of the stochastic model projections as compared to their deterministic counterparts.