P-249 The location of fragments and degraded zones in blastocysts is associated with ploidy: moving towards explaining an AI-based morphology tool trained on euploidy outcomes

Study question

Is the location of degraded areas or fragments an indication of ploidy in blastocyst images?

Summary answer

Degradation traces observed in a blastocyst’s inner cell mass correlates with aneuploidy when confirmed by trophectoderm biopsy.

What is known already

The interaction between humans and Artificial Intelligence (AI) augmented intelligence, (AuI) is dependent on the AI’s ability to be self-explainable and interpretable. This is a highly desired feature of AI’s in healthcare, given that blindly trusting it to make a decision has serious ethical considerations and potential consequences. Currently, most available AI’s provide “black-box” advice that might cause difficult interaction with their human counterparts. ERICA (IVF2.0 Limited, UK), was designed to rank blastocysts using euploid status as ground truth, and although initially a “black-box,” we describe results from an initial attempt towards making it explainable.

Study design, size, duration

This study was designed as a proof-of-concept on retrospectively collected images. De-identified images (n = 329) with known ploidy status (euploid or aneuploid) were retrieved (November 2021) from ERICA. The images were processed from December 2021 to January 2022.

Participants/materials, setting, methods

A senior embryologist identified visual degenerative traces from blastocyst images for areas of cell degradation and cell fragments. Ploidy status was blinded to the embryologist. Images were segmented for trophectoderm (TE), blastocoele (BC), and inner cell mass (ICM) using the automated tool of ERICA’s algorithm. The distance between the centre of each degenerative trace and the ICM was measured. The Dice Similarity Coefficient (DSC) and the proportion of degenerative traces in each zone were computed.

Main results and the role of chance

We identified some level of degradation in 60% of the blastocysts, particularly in BC:44%, ICM:38%, TE:26%, and ICM+BC:55%, and the presence of fragments in 103, particularly in BC:21%, ICM:10%, and TE:24%. Our database contained 52% euploid blastocyst images.

We found that when DSC between degradation and ICM is more than 10% (44/78 aneuploids) the chances of aneuploidy increase by 25% (Z=-1.76, p < 0.05).

We also found a 13% increased chance of an embryo being aneuploid (92/157 aneuploidy) if the area of ICM+BC has any presence of degradation (Z=-1.14, p = 0.13), and an increased risk of aneuploidy if DSC (U = 12401, p = 0.09), and also if the proportion of degradation was found in ICM+BC (U = 12397, p = 0.09).

Our data also suggests that aneuploid embryos have closer fragments (mean=51um, 95% CI: 42.2-59.9) than euploids (mean=63.4um 95% CI:51.1-75.7) (U = 988,=0.19).

Mann-Whitney U test and Z-test for proportions were used accordingly, both under the hypothesis that increased degenerative traces means a higher probability of being aneuploid (one-tailed test).

Limitations, reasons for caution

Analyzing degenerative traces using a single image from a single focal plane might be limiting. Identifying fragments and degradation might not be a replicable process inter- or intra- embryologist. More annotators are needed to reduce this bias.

Wider implications of the findings

Correlation between aneuploidy and cell degradation was stronger in the ICM than TE, although ploidy status is obtained via TE biopsy. Our data suggest that fragments that are closer to the ICM might increase the chances of aneuploidy. A larger prospective multicentre study should be conducted to confirm these findings.

Trial registration number

not applicable

O-235 ERICA (Embryo Ranking Intelligent Classification Assistant) AI predicts miscarriage in poorly ranked embryos from one static, non-invasive embryo image assessment

Study question

Does ERICA’s prognosis ranking based on ploidy, predict early miscarriage following positive biochemical pregnancy test?

Summary answer

The lower ERICA grades embryos, the higher the likelihood of early miscarriage, irrespective of age group.

What is known already

The vast majority of early miscarriages are due to aneuploidy, but preimplantation genetic testing for aneuploidy (PGTA) is potentially invasive, expensive, time-consuming and usually necessitates cryopreservation. Current methods for embryo selection based on morphology and morphokinetics are poorly correlated with ploidy. ERICA is a deep-learning non-invasive tool for embryo ranking, trained to identify ploidy, and has previously been shown to be similar or better than experienced embryologists in assessing implantation potential. AI-based tools capable of embryo ranking and assessment could help save laboratory time and costs, avoiding risk to embryos from invasive techniques.

Study design, size, duration

Retrospective analysis of 599 blastocysts transferred over 12 months in which ERICA was used to assist embryologists during the embryo selection process. ERICA’s prognosis based on ploidy potential is presented as groups labelled as “optimal”, “good”, “fair”, or “poor”. Embryo transfers (ET) reaching biochemical pregnancy (beta-hCG ≥ 20iu) were considered for the study. Early pregnancy loss (EPL) was defined as a biochemical pregnancy failing to develop a gestational sac and/or failure to show heartbeat (FHR).

Participants/materials, setting, methods

ETs resulting in biochemical pregnancies at two IVF clinics were followed-up to FHR till 8 weeks gestation. EPLs were divided into groups according to the presence or absence of a pregnancy sac. ERICA’s suggested prognosis during the embryo selection process was tested against pregnancy outcomes. Further analysis of pregnancy outcomes and their relation to ERICA’s labels was also performed based on age groups. Z-test for two proportions was used to assess statistical significance.

Main results and the role of chance

506 ETs were performed for 599 embryos (mean 1.2 embryos), from which 285 resulted in positive pregnancy tests (56.3%). Thirty-one (10.9%) EPLs happened before the identification of a gestational sac (GS). Ten pregnancies failed to develop FHR after initial GS identification (3.9%), for an overall EPL of 14.4%. The average age in this group was 35.4 years. When evaluated using ERICA’s labels “optimal”, “good”, “fair, and “poor”, chances of miscarriage before GS were 8.9% (8/89); 14.1% (11/78); 18.5% (5/27); and 18.7% (9/48) respectively, where denominator represents total number within a label (i.e. EPL/n). When including all EPLs, chances of miscarriage according to the same labels were 11.2%; 17.9%; 22.2%; and 22.9% respectively.

ERICA’s performance to anticipate the risk of EPL showed statistical significance when the optimal label was compared against all other labels (Z -1.786, p < 0.05), and against the poor prognosis label (Z=-1.653, p < 0.05). After stratifying the dataset according to age groups, increasing miscarriage rates were maintained as ERICA’s prognosis for an embryo worsened, regardless of age groups. The most notable performance was for ≤35-year-olds, where embryos ranked as optimal had an EPL rate of 14.3% in contrast to lowest ranked embryos having a 33.3% EPL rate.

Limitations, reasons for caution

The retrospective nature of this study along with its sample-size might limit the reach of our conclusions, in particular for older patients. The results we present must still be confirmed prospectively, and on a larger dataset.

Wider implications of the findings

Most EPLs are attributed to genetic factors, hence ERICA’s training for embryo ranking was based on ploidy. We conclude that ERICA’s AI is able to identify embryos at a higher risk of EPL non-invasively. Cytogenetic studies from products of miscarriage would help to confirm the hypothesis.

Trial registration number

Not applicable