EVALUATION OF A FULLY AUTOMATED AI TOOL FOR EMBRYO RANKING

NCT07187934 | Recruiting

• 1 other identifier: PRIDE_2025
• Study Type: interventional
• Target: 1,325
• Locations: 1 country
• Sites: 1

Brief Summary

This study aims to investigate whether embryo ranking by a fully automated AI tool (iDAScore) results in a similar transfer-to-pregnancy compared to manual embryo ranking by an embryologist. IVF patients who are planned for a freeze-all approach (where all usable embryos are frozen and transferred in later transfer cycles) will be allocated to either the control (ranking by embryologist) or intervention arm (ranking by iDAScore). The frozen embryos will be warmed and transferred one by one, according to their assigned rank. We want to see if the average number of embryo transfers that is needed to achieve a clinical pregnancy (transfer-to-pregnancy) is similar in both groups.

Conditions

Pregnancy

Keywords

iDA; AI; embryo ranking; transfer-to-pregnancy; time-lapse

Outcome Measures

Primary Outcomes (1)

• Transfer to pregnancy

  Number of embryo transfers performed until a clinical pregnancy is obtained

  The patient is followed up for clinical pregnancy outcomes (7-week echo) up until 1 year after the oocyte retrieval date, or until the result of the last available embryo is known.

Secondary Outcomes (4)

• iDA-embryologist consensus

  Embryo ranking is done on day 5 and/or 6 of embryo development and can be immediately compared during ranking.

• Embryo prioritization time

  Measures the time elapsing between the start and end of embryo prioritisation (range of minutes). Is recorded on day 5 and day 6 of embryo development right after embryo evaluation.

• Cumulative clinical pregnancy rate

  Up until 1 year after oocyte retrieval

• Clinical pregnancy rate after the first embryo transfer

  Clinical pregnancy is determined at the 7 week echo by detection of a fetal heart beat.

Study Arms (2)

Embryologist ranking

NO INTERVENTION

Embryologist ranks the embryos according to their appearance

iDA ranking

EXPERIMENTAL

AI tool ranks embryos based on fully automated analysis of time-lapse images

Device: AI algorithm

Interventions

AI algorithm DEVICE

iDAScore® is a deep learning-based tool that provides fully automated analysis of embryo time-lapse images. It automatically generates a score (1-9.9) for each embryo in the patient's embryo cohort, which correlates to the likelihood of achieving a fetal heartbeat (clinical pregnancy).

Also known as: iDAScore

iDA ranking

Eligibility Criteria

• Age: 18 Years - 42 Years
• Gender Eligibility Details: Only female participants can undergo an oocyte retrieval.
• Healthy Volunteers: Yes
• Age Groups: Adult (18-64)

You may qualify if:

• patients undergoing an ICSI/IVF treatment (with donor or own oocytes) that are planned for a 'freeze-all' strategy on day 5
• all types of sperm samples (ejaculated or testicular, fresh or frozen, partner or donor origin)
• Minimum 8 follicles (≥ 12 mm) on last follicular measurement before oocyte retrieval
• ≤ Age ≤ 42 years
• ≤ BMI ≤ 35 kg/m²
• Maximum third IVF/ICSI cycle
• Participants can be included only once in the trial

You may not qualify if:

• Embassy patients
• Use of frozen oocytes
• In vitro maturation (IVM) cycles
• Pre-implantation genetic Testing (PGT) cycles
• Proven untreated intra-uterine pathologies (e.g. Asherman syndrome, adenomysosis)
• Endometriosis stage 3/4

Sponsors & Collaborators

• Universitair Ziekenhuis Brussel: lead

Study Sites (1)

Brussels IVF, UZ Brussel

Brussels, Brussels Capital, 1090, Belgium

RECRUITING

Related Links

• iDAScore webpage

Study Officials

• Julie Praet, M.D.

  Brussels IVF, UZ Brussel

  PRINCIPAL INVESTIGATOR

Central Study Contacts

Elsie Nulens

CONTACT

+32 2 477 66 48; elsie.nulens@uzbrussel.be

Tine Bosman, Msc

CONTACT

+3224774049; tine.bosman@uzbrussel.be

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: DOUBLE
• Who Masked: PARTICIPANT, CARE PROVIDER
• Purpose: DIAGNOSTIC
• Intervention Model: PARALLEL
• Sponsor Type: OTHER
• Responsible Party: SPONSOR

Model Details: Randomisation (1:1) based on a computer-generated randomisation list with block size 10.

Study Record Dates

• First Submitted: September 4, 2025
• First Posted: September 23, 2025
• Study Start: December 11, 2025
• Primary Completion (Estimated): December 11, 2028
• Study Completion (Estimated): October 1, 2029
• Last Updated: December 23, 2025

Record last verified: 2025-12

Data Sharing

• IPD Sharing: Will not share

Locations

BE (1)