Univfy PreIVF Report Univfy-Powered IVF Refund Program Find a Fertility Doctor Press Room Contact Us About Univfy CEO Presentations
What is your IVF pregnancy success and IVF costs?

About Univfy®

The Univfy® AI Platform makes fertility costs and success more affordable and predictable for women and couples navigating their family-building options. Univfy empowers individuals with personalized information and counseling to inform smarter spending and decision-making, maximizing their chances of success, while increasing growth and efficiency for providers. 

Univfy is the only highly-scalable AI platform to provide scientifically-validated, personalized reports that counsel patients from diverse demographics about their probability of having a baby with IVF. The Univfy® PreIVF Report provides individualized prognostics, showing each patient her own probability of having a baby with up to three IVF cycles. The probabilities are based on the woman or couple’s holistic health profile, including their age, BMI, reproductive history, ovarian reserve test results and their partner’s semen analysis. The predictions are validated based on the clinical IVF outcomes data of the center where the patient is receiving treatment. Univfy’s technology has shown that more than 50% of patients have IVF success probabilities that are higher than estimated by age.

Through technology developed by Stanford University researchers, Univfy uses a rigorous scientific process to develop and validate prediction models and to provide essential information to support patient counselling. Our methods have been published in top, peer-reviewed research publications, in which we reported and established benchmarks for measuring the performance of our IVF outcomes predictive model. Univfy has a global IP portfolio, comprising patents issued in the US and other countries. See below for a timeline of Univfy’s research, partnerships and commercialization.

Univfy Research & Timeline

Our Published Research: Validated IVF Success Prediction Models

About Univfy® Founders

Mylene Yao, CEO and Co-founder, Univfy

Dr. Mylene Yao, Cofounder and CEO, has led Univfy from technology invention and platform development to commercialization. Dr. Yao’s vision is to combine healthcare AI, scientific validation and fintech to power women’s and couple’s access to the most effective and safest fertility treatments. She has more than 20 years of experience in clinical and scientific research in reproductive medicine. Prior to Univfy, Dr. Yao was faculty at Stanford University where she led NIH-funded fertility and embryo genetics research and developed the Univfy technology with the academic founding team.

Dr. Yao graduated from University of Toronto Medical School and completed her OB/GYN residency training at McGill University. She received her clinical subspecialty training in Reproductive Endocrinology and Infertility at Brigham and Women’s Hospital at Harvard University. Dr. Yao received awards for her research on pre-implantation embryo development and uterine receptivity. She is co-author of the Infertility chapter in Berek and Novak’s Gynecology, the top medical textbook for OB/GYNs.

Dr. Yao grew up in Toronto and has made US her home for more than twenty years. She currently lives in the San Francisco Bay Area with her husband and two children.

Professor Wing H. Wong, Cofounder and Scientific Advisor, is the Stephen R. Pierce Family Goldman Sachs Professor in Science and Human Health and Professor in the Departments of Statistics (former chairman) and Biomedical Data Science at Stanford University. Professor Wong and his team began research collaborations with Dr. Yao during the academic research phase of Univfy’s technology development. Professor Wong has provided critical scientific advisory on many fronts, including establishing objective metrics for evaluating prediction model performance and the testing of machine learning parameters required specifically for IVF prediction models. Professor Wong is also a co-inventor on a number of Univfy patent applications.

Professor Wong’s seminal contributions to theoretical statistics – most notably, Bayesian network structures, machine learning, Monte Carlo and applied statistics in computational biology – have been recognized with numerous awards and honors. In 2009, he was elected to the US National Academy of Sciences for developing innovative high-throughput genomics analysis tools, pivotal in driving genomics research in the past decade.

Wing Wong, Scientific Advisor and Co-founder

Univfy® Timeline

Univfy® founding team began their academic research at Stanford University on the application of machine learning (ML) to predict the probability of having a live birth from IVF.

2005

Research: Univfy® founding team reported factors pertaining to embryos as a cohort had greater predictive power in IVF-live birth probabilities than factors describing individual embryos.

2008
In their research, Univfy's founding team identified several powerful predictors of IVF success. Specifically, four variables — the total number of embryos, number of 8-cell embryos, percentage of embryos arrested at cleavage stages, and Day 3 FSH level — were more powerful predictors of IVF success than measures associated with individual transferred embryos. Three of the four predictors above describe the embryo group (or embryo cohort), rather than individual embryos.

Defining human embryo phenotypes by cohort-specific prognostic factors.

Jun SH**, Choi B§, Shahine L, Westphal LM, Behr B, Reijo Pera, Wong WH, Yao MWM. PLoS ONE 2008; 3(7):e2562. doi: 10.1371/journal.pone.002562. **Co-first authors.

Univfy® co-founders received The Stanford-Coulter Translational Research Program Award.

2009

Dr. Mylene Yao and Professor Wing H. Wong co-founded Univfy®.

2009

Research: Using a prediction model developed and validated with machine learning, more patients were found to have higher IVF success probabilities than age-only control model.

2010
More than 80% of patients tested by Univfy’s prediction models were proven to have higher probabilities of IVF success, compared to age-based model.

Further, nearly 60% of patients were found to have a better chance of having a baby than was estimated by age.

Why the discrepancy? Because age-based predictions fail to consider important reproductive factors that more reliably predict IVF success. Likewise, other methods that analyze fertility factors independently do not consider the patient’s entire reproductive profile either. Univfy IVF Prediction Tests analyze an entire profile of reproductive factors to deliver the most accurate estimate of having a baby, whether from a first IVF or subsequent IVF treatment.

Univfy IVF prediction models have 1,000 times bigger likelihood than age-based estimates. Alternatively, we measured the percentage improvement in predictive power by comparing the posterior log-likelihoods of each prediction model and its age-based prediction, against the baseline model. Univfy PreIVF shows a 36% improvement in predictive power over age-based prediction, while Univfy PredictIVF shows a 77% improvement in predictive power over age-based prediction.

Deep phenotyping to predict live birth outcomes in in vitro fertilization.

Banerjee P**, Choi B§, Shahine LK, Jun SH, O’Leary K, Lathi RB, Westphal LM, Wong WH, Yao MWM. PNAS 2010;107(31):13559-60. doi: 10.1073/pnas.1002296107. Epub 2010 Jul 19. **Co-first authors.

Univfy® raised its first seed round and began its R&D* as a company.

2010

Product: Univfy® provided a validated IVF outcomes prediction tool called IVFSingleTM to support providers in their eSET counseling.

2012

Research: Using machine learning, we showed that the probability of having twins from transferring two embryos in an IVF-embryo transfer procedure can be predicted for each patient and her partner.

2012
Univfy has developed and tested a prediction model that can calculate a woman’s probability of having twins from IVF (when two embryos are transferred). The model uses a woman’s reproductive history, laboratory tests, response from her current IVF treatment, and embryo quality to determine her probability of having twins if two fresh embryos are transferred instead of just one. Personalized predictions of multiple-birth risk can support physicians when counseling their patients about the benefits of elective single embryo transfer (eSET).

Predicting personalized multiple birth risks after in vitro fertilization-double embryo transfer.

Lannon BM**, Choi B§, Hacker MR, Dodge LE, Malizia BA, Barrett CB, , Wong WH, Yao MWM, Penzias AS. Fertil Steril 2012;98(1):69-76. doi:10.1016/j.fertnstert.2012.0411. Epub 2012 Jun 4. **Co-first authors.

Product: Univfy® launched the Univfy PreIVFTM Report* to support provider-patient counseling.

2013

Research: We reported that first-cycle IVF data and outcomes from multiple centers can be merged to train and validate a multi-center IVF-live birth prediction model that has superior predictive power, discrimination and reclassification rates than possible from an age-only control mode.

2013
We have found that patient data can be used to accurately predict a woman’s probability of IVF success from a first cycle. Univfy PreIVF analyzes each individual’s fertility profile — including age; body mass index, (BMI) comprising height and weight; ovarian reserve tests (e.g. serum anti-mullerian hormone or Day 3 FSH tests); semen analysis; and prior fertility and medical history. We then compare that information to profiles of thousands of others who have completed first IVF cycles to deliver a personalized prediction of IVF success instantly. More than half of first-time IVF cases tested by Univfy PreIVF had a significantly higher likelihood of IVF success than estimated by age.

Personalized prediction of first-cycle in vitro fertilization.

Choi B, Bosch E, Lannon BM, Leveille MD, Wong WH, Leader A, Pellicer A, Penzias AS, Yao MWM. Fertil Steril 2013;99(7):1905-11. doi: 10.1016/j.fertnstert.2013.02.016. Epub 2013 Mar 21.

Research: Univfy® and collaborators show the ovarian reserve marker AMH, when used together with other clinical predictors, provided predicted probabilities of IVF success with or without the concomitant use of antral follicle count.

2015

Antimüllerian hormone levels and antral follicle count as prognostic indicators in a personalized prediction model of live birth.

Nelson SM, Fleming R, Gaudoin M, Choi B, Santo-Domingo K, Yao MWM. Fertil Steril 2015. doi: 10.1016/j.fertnstert.2015.04.032

Product: Univfy® AI Platform launched, integrating the Univfy PreIVF ReportTM and Univfy-Powered IVF Refund Program*

Univfy integrates clinical and financial counseling to support physician-patient counseling. Physicians counsel patients with the Univfy PreIVF Report to give patients their personalized probabibility of IVF success. The Univfy-powered IVF Refund Program gives patients the ability to afford multiple IVF cycles, if needed.

2016-2018

Univfy® is recognized as a rising leader in healthcare technology.

2016-2017
  • 2016: Univfy voted winner in Consumer Electronic Show 2016 The Bump Best of Baby Tech Awards in the Fertility category
  • 2016: Univfy CEO and Cofounder Dr. Mylene Yao honored at The Bump Moms: Movers + Makers Awards Luncheon, Los Angeles
  • 2016: Univfy selected to join Springboard Tech Hub 2016
  • 2017: Dr. Mylene Yao honored among MM&M's Top 40 Healthcare Transformers

Univfy® Raises Series A Funding

2018

Product: Univfy® Expands AI Platform Services

2018-2019
  • Univfy® Business Analytics for Providers
  • Univfy® Patient CRM for patient retention
  • Univfy® Egg Freezing Report
  • Univfy® PreIVF Report for IVF Using Donor Egg

Partnership: Univfy® and BBVA USA collaborate to bring Express Healthcare Loan Program to fertility patients.

2019

Product: Univfy® Launches New AI Platform Services

2020
  • Your Fertility® Report to increase patient retention
  • Univfy® Interactive Ad Campaign for consumers
  • Univfy® Fertility Counseling Concierge
  • Univfy®-Run-Reports

Research: Univfy® Collaborates with EU Fertility researchers to discuss the use of AI-assisted IVF prognostics as one approach to reducing barriers to IVF utilisation.

Economic factors and stress are among key barriers to ART utilization. One patient-centred approach to addressing these barriers is to clearly communicate individualized prognostic information to patients with transparency and empathy, thereby helping patients set realistic expectations of ART [assisted reproductive technology]. Shared decision-making may also reduce the stress of healthcare practitioners counselling patients on ART prognosis.

2020

Empathetic application of machine learning may address appropriate utilisation of assisted reproductive technologies.

Jenkins J, van der Poel S, Krüssel J, Bosch E, Nelson SM, Pinborg A, Yao MWM. Reproductive Biomedicine Online 2020. doi:10.1016/j.rbmo.2020.07.005