fbpx
Our Impact
30 August 24

Abstract (CBRA 2024): Correlation of oocyte quality assessment by artificial intelligence to blastocyst development, and ovarian stimulation protocols

Braga, D., Setti, A., Fjeldstad, J., Mercuri, N., Mojiri, P., Iaconelli Jr., A., Borges Jr., E.
OBJECTIVE:
MAGENTA™ is an artificial intelligence (AI) tool that assesses images of mature denuded oocytes to provide an analysis shown to significantly correlate with subsequent blastocyst development and quality, consistently outperforming embryologists in this task. MAGENTA™ reports provide an objective measure of oocyte quality, including images and individualized scores (0-10) that predict blastocyst formation for each oocyte. Therefore, this AI tool can be extremely useful not only for predicting embryo implantation potential but also for improving embryo selection for transfer, facilitating the management of patient expectations, and allowing the transfer of fewer embryos, thereby reducing multiple pregnancies and associated risks. Furthermore, it can help us better understand the biology of the oocyte and its response to different protocols of controlled ovarian stimulation, and pituitary blockage. Therefore, the objectives for this study were to investigate: (i) whether MAGENTA™ oocyte score (MS) may predict blastocyst development and morphokinetic factors, (ii) the influence of pituitary blockage on oocyte quality, assessed by MAGENTA™ and, (iii) to evaluate the influence of the controlled ovarian stimulation (COS) protocol on oocyte quality, assessed by MAGENTA™.
 
METHODS:
For this study, MS was applied to 7,783 oocyte images generated by the time-lapse imaging system (TLI). For the primary objective, oocytes were split into two groups according to the blastocyst development outcome: Blastocyst (n=4,612) and Non-blastocyst Group (n=3,171) and embryo morphokinetic data was analyzed. For the secondary objective, oocytes were split into groups depending on the protocol used to prevent the LH surge: progestin-primed (n=707) and GnRH-antagonist group (n=5,537) and, for the third objective, (iii) oocytes were divided into groups based on the COS protocol: FSH-only (n=2,270) or FSH + LH (n=2,879). The correlation of MS to blastocyst formation and morphokinetic data: timing to pronuclei appearance and fading (tPNa and tPNf), two (t2), three (t3), four (t4), five (t5), six (t6), seven (t7), and eight cells (t8), morulae (tM), start of blastulation (tSB) and blastulation (tB), was evaluated. In a second analysis, the correlation of the pituitary blockage and COS protocol to MS were evaluated.
 
RESULTS:
The MS was significantly higher in oocytes that reached the blastocyst stage when compared to those that did not achieve the blastocyst stage (6.5 vs 5.5, p<0.001 Welch’s t-test). A negative correlation between MS and morphokinetic factors was noted (indicating shorter durations between developmental stages is favorable), especially for early embryos divisions (tPNa, tPNf, t2, t4, t6, t7, and t8). No differences were observed on MS when different protocols were used to prevent the LH surge (p=0.191 Welch’s t-test). When COS protocols were compared, no significant differences were noted when the whole group was evaluated (p=0.225), but when the oocytes were divided by age (<35 and ≥35), a higher MS was observed in oocytes derived from cycles stimulated with FSH + LH (5.9) when compared with dose stimulated exclusively with FSH for the ≥35 age group (5.6, p< 0.01 Welch’s t-test).
 
CONCLUSIONS:
In this study, an AI model for predicting blastocyst formation was validated. The correlation between the MS and morphokinetic parameters adds a layer of confidence to the model. It’s reassuring since the connection between morphokinetic data, and the embryo implantation potential is already known. Additionally, our findings suggest that older patients might benefit at the oocyte level from the addition of LH in the COS. This corroborates with previous studies indicating that embryos from cycles stimulated with LH tend to develop faster and have an increased chance of implantation, especially for women of advanced age.
 

> Return to our full library of research studies

STAY IN THE LOOP

Join our mailing list for dispatches on the future of fertility

    Your personal information will be used to respond to your request and will be governed by our Privacy Policy.

    Form was submitted!