Email
Password
Remember meForgot password?
    Log in with Twitter

article imageArtificial intelligence to aid IVF treatment: Interview Special

By Tim Sandle     Dec 1, 2017 in Health
The company Univfy is using artificial intelligence and predictive technology to help women more accurately know their chances of IVF success. To find out more we spoke with Mylene Yao, the company's CEO.
Univfy analyzes many factors in a patient’s fertility profile, such as age, body mass index, ovarian reserve test results, semen analysis, and clinical diagnoses. This is in order to give personalized probability of in vitro fertilisation (IVF) success. A secondary aim is to bring more transparency to the success and cost of IVF.
Moreover, by using Univfy patients can learn about the costs of IVF treatment and their probability of success after one, two or three IVF cycles. The platform, therefore, also helps couples to make better financial decisions.
To discover more about this combination of artificial intelligence and predictive technology, we spoke with the Chief Executive Officer, Dr. Mylene Yao.
Digital Journal: Thanks for the interview. What are the established methods for couples seeking to have children?
Dr. Mylene Yao: First, about 85 percent of couples in their 20s conceive naturally over a year if they're trying every month. You can increase your chance of conceiving by timing sex around and in the 36 hours after ovulation.
Women who have risk factors such as past sexually transmitted diseases, irregular periods, certain birth defects, or medical problems that can impact fertility (like thyroid disease) should consult a fertility specialist regardless of her age, to determine if her fertility may be compromised and whether a specific treatment would improve her chances of having a baby.
Men who have risk factors such as undescended testes and past infection that can impact sperm production (such as mumps orchitis) should also consult a fertility specialist or urologist/andrologist. Often, the doctor will order a semen analysis to determine if there is male risk factor.
If the fertility doctor confirms that there are risk factors and the couple's chances of natural conception is low, then he/she would normally recommend one or more treatment options. The exact treatment will depend on the cause of subfertility (a term used in the U.K.) or infertility (a term used in U.S.). For most couples, in vitro fertilization (IVF) is a highly effective medical treatment that significantly improves the chances of having a baby.
DJ: What alternative methods are there?
Yao: The treatment methods depend on the cause of infertility. Also, the age of the couple -- especially the age of the woman -- should be taken into account when deciding on the treatment method. As a woman ages, her ovarian reserve decreases, so if IVF is the most effective treatment for a women, you would not want to delay IVF treatment because the same treatment can give lower results if it is done later in life.
Some commonly used methods of fertility treatment (not necessarily "alternatives" as they are not applicable in every patient) are ovulation induction (if irregular ovulation is the only cause), intrauterine insemination (usually done as an alternative to IVF due to its lower cost), tubal surgery (if tubal factor were the only cause and the woman has good ovarian reserve), the use of donor sperm (if there is male factor infertility).
The Play of Light and Gray Over a Pregnancy
A pregnant woman
Hugrakka
DJ: How does Univfy work?
Yao: Univfy is a predictive analytics platform that combines machine learning, artificial intelligence, and fintech to improve the patient's experience. Most importantly, we help more women to succeed in having a baby from IVF by working with their doctors to help them afford several IVF cycles.
Many more patients can have a baby if they could afford to do more than one IVF treatment. Without the use of machine learning, much fewer patients actually qualify for refund warranty programs.
Univfy uses an IVF center's own past data and outcomes to develop and validate a prediction model of IVF success. We then integrate the IVF success prediction model with the clinic's IVF fee schedule to help fertility doctors offer special pricing programs that are designed to help more women afford several IVF cycles, if needed.
DJ: How does Unify help with the cost of IVF?
Yao: Univfy helps to connect the patient's financial cost and her specific probability of having a successful treatment. Doctors using the Univfy platform report that the patient's experience is dramatically improved by the level of transparency and accountability. Doctors want to provide transparency but in the past, without the Univfy PreIVF Report, it was technically not possible.
With the Univfy PreIVF Report, doctors can provide much more personalized counseling when supported by the Univfy PreIVF Report, which tells a patient her probability of having a baby from one, two, or three IVF treatments, the probability of not having a baby even after 3 tries, and the financial options that are available. Traditionally, medical counseling and financial counseling are conducted separately. However, patients who have to pay for IVF out-of-pocket need to understand their medical and financial risks and benefits in order to make an informed decision.
A pregnant woman
A pregnant woman
David Roseborough / Creative Commons
DJ: How was the artificial intelligence developed?
Yao: The Univfy technology was originally developed by myself and Professor Wing H. Wong, at Stanford University as part of an academic research project. Univfy has licensed the technology and also developed new technology during its commercialization efforts, all under the Univfy global IP portfolio.
DJ: Which factors are assessed to help assess IVF success?
Yao: The Univfy PreIVF Report uses health data that are available from standard clinical evaluation by fertility doctors. Health data such as age, body mass index, ovarian reserve test results, clinical diagnosis, reproductive history, past use of fertility treatment and results, and semen analysis are used by the prediction model to compute the predicted probability of IVF success.
DJ: How have you assessed the predictive technology? Did you run trials?
Yao: Each prediction model that we provide to fertility doctors has been subjected to rigorous testing and validation, even much more than examples that we have published in top, peer-reviewed clinical research journals.
Since we are providing predictive technology to patients who have already been determined to have infertility and for whom doctors are already recommending IVF treatment. Univfy does not make the diagnosis or recommend the treatment), clinical trials are not applicable. Our goal is to help patients to be more informed about the IVF treatment that has been recommended to them and to be able to afford IVF, should they decide to take their doctor's recommendations.
DJ: What has been the reception from the medical profession?
Yao: Fertility doctors are very happy with the Univfy PreIVF Report because the report has dramatically improved their patient's experience. They feel that they can provide more personalized counseling and the level of transparency and accountability that their patients deserve.
DJ: How have patients reacted?
Yao: Patients very much appreciate having a very data-driven, validated report that is personalized to their own health data. We have heard very positive feedback from doctors using the Univfy PreIVF Report.
DJ: Which other areas are you working on?
Yao: Currently patients experience a huge financial barrier when seeking help from IVF treatment. We are working to expand our financial program to incur cost savings on medication and genetic testing so that more patients can access IVF and utilize genetic testing technology, which can be very beneficial for some patients. We're hoping to develop a consumer loan program to help patients get their needed treatment sooner rather than later.
More about Ivf, Artificial intelligence, Predictive technology, Fertility
More news from
Latest News
Top News