SAN FRANCISCO BAY AREA, Calif., April 18, 2025 /PRNewswire/ -- Univfy, a pioneering force in the realm of fertility and health artificial intelligence, has unveiled groundbreaking research that showcases the impressive performance of its IVF live birth prediction models. This research, published in the esteemed journal Nature Communications, reveals a significant advancement in clinic-level, real-world validation of Univfy's proprietary models. These models play a critical role in addressing economic and clinical challenges associated with in vitro fertilization (IVF), ultimately enhancing access, affordability, and outcomes for patients seeking fertility treatments.

The study, titled "Machine learning center-specific models show improved live birth predictions over US national registry-based model," demonstrates that Univfy's models consistently outperformed the US national registry model in several key metrics. Notably, metrics such as the F1 scorean essential measure combining precision and recalland the precision-recall area-under-the-curve (PR AUC) highlighted the superior accuracy of Univfy's predictions. These enhancements are vital because they directly correlate to reducing false positives and false negatives, a factor that is indispensable for developing effective economic solutions like value-based care and actuarial models.

In a comprehensive analysis involving 4,645 patients across six fertility centers, the Univfy models accurately predicted that 76% of the patients had a first-cycle IVF live birth probability (LBP) of 50% or greater. Alarmingly, it was found that 23% of these patients, who were forecasted to have LBPs of 50% or more by Univfy, received lower predictions from the US national registry model. Furthermore, Univfy's models correctly identified that 11% of patients had LBPs of 75% or higher, a prediction that was not recognized by the national model, which failed to identify any patients within this category despite an actual live birth rate of 81%.

This publication stands as a significant milestone in the validation of the science underpinning the Univfy AI/ML Platform. It emphasizes the platform's potential to foster economic solutions, such as value-based IVF care, essential for broadening access and affordability of IVF treatments. This research builds on prior findings indicating that utilizing Univfy's PreIVF Report for patient counseling can lead to a 2-3 times increase in IVF utilization across centers in the United States, United Kingdom, and the European Union. Such results further underscore the importance of locally validated AI-powered solutions in real-world clinical contexts.

Dr. Mylene Yao, CEO and Co-founder of Univfy, expressed her enthusiasm regarding the publication, stating, "This publication is a testament to the rigorous science behind the Univfy AI/ML Platform, enabling it to correctly predict the excellent IVF outcomes achieved by our collaborators and the broader IVF ecosystem that are often underappreciated. Univfy was established with the mission to improve patient-centric care, particularly in supporting IVF prognostic counseling. Our efforts in enhancing IVF cost-success transparency have led us to create a platform that facilitates the scalable production of validated economic solutions. This is not only beneficial for patients and providers but is also urgently needed to help more women and couples access and afford IVF to build their families."

The findings of this study yield significant advantages for various stakeholders in the fertility treatment landscape:

  • For Patients: Enhanced IVF cost-success transparency enables informed decisions regarding fertility care and family planning.
  • For Healthcare Providers: Improved patient counseling coupled with streamlined clinical workflows enhances efficiency, reducing delays in treatment caused by the underestimation of IVF efficacy.
  • For Health Insurers and Benefits Programs: Better member experiences along with increased transparency and support for providers facilitate IVF coverage expansion while achieving cost savings.

About Univfy
Univfy, a Series B company headquartered in the San Francisco Bay Area, is dedicated to enhancing IVF success rates while making treatments more accessible and affordable. Founded by researchers from Stanford University, the proprietary Univfy AI/ML Platform delivers accurate, personalized, and validated probabilities of having a baby through IVF and other treatment options. This empowers patients to make informed decisions about their fertility journeys. Additionally, the Univfy platform allows providers to offer scalable value-based IVF pricing, fostering affordability. With a B2B model, Univfy has successfully commercialized provider solutions in the United States, United Kingdom, and European Union, supporting providers in patient counseling and clinical analytics to improve patient-centric care.

About Infertility, IVF and the IVF Market
Infertility affects approximately one in six individuals of reproductive age, translating to over 200 million people globally. In the United States alone, around 7-10 million people struggle with infertility. Annually, roughly 4 million IVF treatments are conducted worldwide, resulting in over a million births. Despite the safety and efficacy of IVF, many patients face significant barriers, including high costs, limited insurance coverage, and uncertain success rates, leading to a mere 3% utilization rate in the US.

Recent estimates indicate that the global IVF market is valued at around USD $25 billion in 2023 and is projected to reach approximately USD $44 billion by 2033, growing at a compound annual growth rate (CAGR) of 5.57% during this period.

Media Contact:
media@univfy.com
info@univfy.com

SOURCE Univfy Inc.