AI-Driven Model Impact on Patient Engagement in Medically Assisted Reproduction

Instituto Valenciano de Infertilidade de Lisboa

Status: Enrolling

Conditions: Infertility (IVF Patients); Artificial Intelligence (AI)

Treatments: Other: Artificial intelligence-Machine learning report with accurate personalized probabilities of having a live birth rate

Study type: Observational

Funder types: NETWORK; Industry

Identifiers

NCT07087171

2412-LIS-233-AN

Details and patient eligibility

About

Infertility is a globally significant medical condition, profoundly impacting individuals and couples both emotionally and physically. The multifaceted nature of in vitro fertilization (IVF) treatment demands active patient participation, with engagement playing a pivotal role in treatment success and satisfaction. However, suboptimal engagement can lead to challenges such as not initiating treatment, missed appointments, medication errors, dropping out and heightened stress levels, all of which may adversely affect clinical outcomes.

Recent advancements in Artificial Intelligence (AI) and Machine Learning (ML) have revolutionized healthcare, offering innovative solutions for personalized patient care. In IVF, AI-ML models hold the potential to enhance patient engagement by delivering tailored communication, reminders, and educational support, but also improved prognostication by providing personalized and accurate predictions of treatment outcomes. These capabilities enable patients to make more informed decisions and enhance their adherence to treatment protocols.This protocol outlines a prospective evaluation of an AI-ML model, specifically the Univfy PreIVF report, developed to improve patient engagement in IVF care. Recently, a retrospective, multicenter study reported improved IVF utilization rates among patients counselled using the Univfy PreIVF Report. The current study will prospectively assess the model's effectiveness in addressing individual patient needs and creating a supportive treatment environment. Specifically, this study will measure adherence to providers' recommendation of treatment protocols. By analyzing the impact of these interventions, this research aims to provide robust evidence for the integration of AI-ML technologies in reproductive medicine, paving the way for broader implementation and improved patient outcomes.

Enrollment: 774 estimated patients

Sex: All

Ages: 18 to 45 years old

Volunteers: No Healthy Volunteers

Inclusion criteria

• Infertile patients aged 18-45 years
• Patients willing to undergo Medically Assisted Reproduction (heterosexual couples, same-sex female couples and single females undergoing artificial insemination, IVF/ICSI or oocyte donation treatments)

Exclusion criteria

• Age >45 years
• Patients who are not candidates for IVF/ICSI
• Patients who are menopausal or peri-menopausal
• Patients undergoing Fertility Preservation
• Same-sex couples who will undergo reception of oocytes from partner.
• Patients who decline to be counselled about their probability of having a live birth from IVF/ICSI treatment

Trial design

774 participants in 2 patient groups

Conventional counselling group

Description:

A retrospective cohort of patients who underwent their new patient visit with one of the doctors participating in the study between December 2024 and June 2025 will be analyzed.

AI-based counselling group

Description:

A prospective cohort of patients undergoing their NPV with one of the doctors participating in the study will receive an artificial intelligence-machine learning report with their accurate personalized probabilities of having a live birth rate together with a medical explanation by their physician

Treatment:

Other: Artificial intelligence-Machine learning report with accurate personalized probabilities of having a live birth rate