AI Revolutionizes Reproductive Medicine

Through this article, Joseph Chervenak, MD/MBA, Reproductive Endocrinology and Infertility Fellow at Montefiore Health System, shares his insights on the diverse and impactful field of Obstetrics and Gynecology. He reflects on his extensive training, the influence of his mentors and the interdisciplinary nature of the specialty. Chervenak emphasizes the transformative advancements in reproductive medicine, including IVF, cryopreservation and AI applications in fertility care. He highlights the balance between innovation and empirical practice and discusses the future potential of AI to improve access and quality of care in reproductive medicine.

During residency interviews when a medical student is asked why they want to go into Obstetrics and Gynecology or OB/GYN, a common response is that they get “to do a little bit of everything,” as the field encompasses the full range of healthcare practice from outpatient primary care to surgery.

After completing 7 years of post-medical school graduate training, I can attest to the breadth and reach of the field in the pursuit of women’s health. There is something about the scope of experiences that training provides that makes reproductive medicine physicians some of the best patient advocates I know. I have had the privilege of training with some truly outstanding minimally invasive surgeons, highrisk pregnancy experts and oncologists.

 Inspired by my parents—both maternal-fetal medicine physicians—I always knew that the field had much to offer in the way of being able to help patients at some of the most important times in their lives. Pursuing training at Cornell, I was mentored by some of the best in their field.

In a healthcare landscape defined by siloed specialities, the field seemed to offer the benefit of cutting-edge subspecialty practice and interdisciplinary collaboration. In this way, the discipline offers the unique possibility of being able to apply ever more specialized innovation in a way that can positively impact the lives of many.

 I have been fortunate to receive fellowship training in Reproductive Endocrinology and Infertility, a field that has been transformed in recent years by a clear societal need and great advancements in the success of fertility-related treatments. The fundamental treatment of assisted reproduction, In Vitro Fertilization (IVF), is more effective and accessible than ever.

Cryopreservation or egg freezing has meaningfully enhanced family-building options and created invaluable flexibility for patients. Related procedures such as preimplantation genetic testing (PGT) hold tremendous promise. As elsewhere in healthcare, the challenge of developing these pioneering technologies while maintaining affordability and accessibility remains acute. New developments have the potential to realize this critical goal.

AI has become synonymous with the future of the healthcare industry and just about every industry. I have been fascinated by the field’s progression ever since I studied neuroscience as an undergraduate at Columbia. The black box of the human brain was an inspiration for some of the deep learning models powering the impressive feats we see today. One of the remarkable developments I have seen through my training is for powerful AI models to find such meaningful applications in healthcare. This has special potential in fertility, where models have already proven effective at technical tasks such as embryo selection, IVF cycle management and even patient counselling.

I am proud to have been part of the first publications to evaluate the effectiveness of large language models (LLMs) in the field of OB/GYN and fertility care respectively. We presented work at the American Society for Reproductive Medicine (ASRM) that highlighted the potential for LLMs to improve the readability of patient counselling materials rapidly and effectively. I believe that these successes can help make meaningful strides in improving access to care without sacrificing quality.

Of course, a healthy scepticism towards new treatments is still needed. Medicine is a field that is driven by practice informed by decades of empirical knowledge. The training we receive during medical school, residency and beyond imbues us with accumulated lessons learned from what works and what does not.

Physicians have a high threshold to adopt new technology as ‘first, not harm’ means that new approaches need to establish a favourable risk/benefit profile before responsibly being put into practice. This balance is paramount in a field that can influence entire lives. Ensuring that everything goes right at the delivery of a newborn, for example, has tremendous lifelong importance. Despite these stakes, the field’s strong record of implementing innovation suggests that the future is bright.

Over the past few decades, reproductive medicine has successfully incorporated technologies ranging from ultrasound to IVF which have unlocked new possibilities in diagnosis and treatment. I believe AI and related developments have similar potential to usher in an exciting new era in our field.

As our society grapples with demographic changes, political challenges and limitations in access to care, reproductive medicine and fertility will remain a crucial area of practice. Lessons from our field have the potential to inform how to meet challenges and continue to improve healthcare more broadly