AI-TOP Study Artificial Intelligence for Trigger Optimization.

Официальное название

An Artificial Intelligence Based Approach for Selecting the Optimal Day for Triggering.

Обзор

To evaluate, in a randomized controlled trial, whether AI-guided monitoring and ovulation triggering leads to clinical outcomes comparable to those achieved through physician-led decision-making in patients undergoing ovarian stimulation for IVF.

Подробное описание

Assisted Reproductive Technology is undergoing a major transformation with the introduction of artificial intelligence (AI), which is reshaping how medical treatments are carried out. In IVF, one of the persistent challenges has been maximizing the number of oocytes retrieved while efficiently managing clinical workload-particularly by reducing weekend procedures-without compromising outcomes. Although a patient's response may vary between cycles, evidence shows that adjusting the trigger day by one day does not significantly affect clinical results, enabling more flexible scheduling.

AI enables a shift from standardized protocols to personalized treatments, improving clinical outcomes, streamlining processes and enhancing operational efficiency. Recent research shows that AI-based models can optimize ovarian stimulation, improve trigger-day selection, and increase the number of fertilized oocytes compared to decisions made solely by physicians. AI algorithms have also accurately predicted the number of oocytes retrieved, contributing to more effective protocols and higher live birth rates.

Beyond trigger timing, AI has been shown to improve workflow efficiency in IVF clinics by optimizing monitoring schedules and balancing clinical workload without negatively affecting cycle outcomes.

Based on this growing evidence, a randomized controlled trial was designed to compare clinical outcomes of controlled ovarian stimulation when trigger and retrieval decisions are made solely by the physician versus when the physician is assisted by AI guidance.

Вмешательства

• Устройство STIMAI®.
  The AI algorithm used in this study is STIMAI®. STIMAI® is an artificial intelligence-based software that assists clinicians by providing data-driven insights to optimize the fertility treatment process and support conception. The software is designed as a clinical decision support tool and does not replace the physician's judgment; final clinical decisions will remain under the responsibility of the treating physician The physician will consult the AI application, which predicts the number of M
• Другое Routine clinical management
  As soon as 2-3 follicles of 17 mm are detected, the physician will determine the timing of ovulation triggering based on clinical judgment.

Первичные конечные точки

• MII oocytes [Срок оценки: Day of pickup approx. 34-36 hours after ovulation trigger.]

Вторичные конечные точки (7)

• Distribution of retrieval procedures during the week. [Срок оценки: Assessed at the end of the stimulation cycle, once the retrieval schedule is completed. approx. 34-36 hours after ovulation trigger.]
• Number of COCs [Срок оценки: Measured on the day of oocyte retrieval. approx. 34-36 hours after ovulation trigger.]
• Length of stimulation (days) [Срок оценки: From the first day of stimulation until the day the ovulation trigger is administered. Up to 8-15 days]
• FORT (pre-ovulatory follicles on trigger day/AFC) [Срок оценки: AFC is measured at baseline (cycle day 2-3); pre-ovulatory follicles are counted on trigger day. Up to 8-15 days]
• FOI (N COCs/AFC) [Срок оценки: AFC measured at baseline; COCs counted on the day of retrieval approx. 34-36 hours after ovulation trigger.]
• Number of visits [Срок оценки: Counted from the start of stimulation until the trigger day. up to 8-12 days]
• Spontaneous ovulation [Срок оценки: Detected between the trigger administration and the planned oocyte retrieval. approx. 34-36 hours after ovulation trigger.]

Критерии участия

Критерии включения

• Undergoing COS for IVF with autologous oocytes, oocyte donation and elective fertility preservation with all monitoring USS (ultrasound scan) conducted at our centers.

Критерии исключения

• Medically indicated fertility preservation
• Inability to attend clinic visits for monitoring.

Критерии приведены из реестра в оригинале (на английском). Окончательную оценку соответствия проводит исследовательский центр.

Здоровые добровольцы: Нет

Дизайн исследования

Распределение

Рандомизированное

Модель

Параллельные группы

Маскирование

Открытое

Основная цель

Лечение

Центры проведения

Испания · 5 центров

• Dexeus Mujer Sabadell — Sabadell
• Dexeus Mujer Sant Cugat — Sant Cugat del Vallès
• Dexeus Mujer Reus — Reus
• Hospital Universitario Quiron Dexeus — Barcelona
• Dexeus Mujer Tarragona — Tarragona

Публикации

• Babayev E. Man versus machine in in vitro fertilization-can artificial intelligence replace physicians? Fertil Steril. 2020 Nov;114(5):963. doi: 10.1016/j.fertnstert.2020.07.042. Epub 2020 Aug 17. No abstract available. PMID 32819677
• Blockeel C, Engels S, De Vos M, Haentjens P, Polyzos NP, Stoop D, Camus M, Devroey P. Oestradiol valerate pretreatment in GnRH-antagonist cycles: a randomized controlled trial. Reprod Biomed Online. 2012 Mar;24(3):272-80. doi: 10.1016/j.rbmo.2011.11.012. Epub 2011 Nov 30. PMID 22296973
• Canon C, Leibner L, Fanton M, Chang Z, Suraj V, Lee JA, Loewke K, Hoffman D. Optimizing oocyte yield utilizing a machine learning model for dose and trigger decisions, a multi-center, prospective study. Sci Rep. 2024 Aug 20;14(1):18721. doi: 10.1038/s41598-024-69165-1. PMID 39164339
• Chow DJX, Wijesinghe P, Dholakia K, Dunning KR. Does artificial intelligence have a role in the IVF clinic? Reprod Fertil. 2021 Aug 23;2(3):C29-C34. doi: 10.1530/RAF-21-0043. eCollection 2021 Jul. PMID 35118395
• Dimitriadis I, Zaninovic N, Badiola AC, Bormann CL. Artificial intelligence in the embryology laboratory: a review. Reprod Biomed Online. 2022 Mar;44(3):435-448. doi: 10.1016/j.rbmo.2021.11.003. Epub 2021 Nov 12. PMID 35027326
• Ferrand T, Boulant J, He C, Chambost J, Jacques C, Pena CA, Hickman C, Reignier A, Freour T. Predicting the number of oocytes retrieved from controlled ovarian hyperstimulation with machine learning. Hum Reprod. 2023 Oct 3;38(10):1918-1926. doi: 10.1093/humrep/dead163. PMID 37581894
• Hariton E, Chi EA, Chi G, Morris JR, Braatz J, Rajpurkar P, Rosen M. A machine learning algorithm can optimize the day of trigger to improve in vitro fertilization outcomes. Fertil Steril. 2021 Nov;116(5):1227-1235. doi: 10.1016/j.fertnstert.2021.06.018. Epub 2021 Jul 10. PMID 34256948
• Letterie G, MacDonald A, Shi Z. An artificial intelligence platform to optimize workflow during ovarian stimulation and IVF: process improvement and outcome-based predictions. Reprod Biomed Online. 2022 Feb;44(2):254-260. doi: 10.1016/j.rbmo.2021.10.006. Epub 2021 Oct 20. PMID 34865998

Идентификаторы

NCT: NCT07515118 · FSD-AIG-2025-20

Первоисточники (государственные реестры)

• ClinicalTrials.gov (NIH, США)

Открыть это исследование на ClinicalTrials.gov ↗