The impending AI revolution of reproductive medicine
Artificial intelligence could improve outcomes, but some researchers worry about pitfalls
Artificial intelligence has been in the news because of its impact on entertainment, politics, and other endeavors. In the field of reproductive medicine, researchers are on the precipice of a revolution in care. But there are potential pitfalls with the introduction of AI into the field as there are with other industries.
Fertility and Sterility, the journal of the American Society for Reproductive Medicine, recently published a series of articles about how the technology would best be introduced and utilized by reproductive healthcare professionals. Dr. Marcelle Cedars, the immediate past-President of ASRM, spoke to Repro Rights Now about where things could develop going forward.
“There are a number of avenues where artificial intelligence can be helpful,” Cedars said. “One of the benefits of artificial intelligence is the ability to process and combine data in definitely more efficient ways than the human brain can much faster.”
Artificial intelligence would help IVF professionals individualize treatment to make it more effective based on the medical conditions that a person has. It could make better predictions about which type of fertility treatment would work best. Currently, they put patients into groups by age.
Embryo selection is where artificial intelligence is the most advanced. It’s largely selective when IVF providers select an embryo. They look at images from time-lapse machines to decide. There is so much more an AI machine could consider, including the age, weight, and prior obstetrical history of a patient when selecting an embryo.
“Right now, it's really focusing on just what the camera sees in the laboratory, which is equivalent to the human eye,” Cedars said. “But with the thought that the AI interpretation of those images will be able to process more images and make better decisions.”
Currently, IVF treatment can be costly. The goal of any evolution in medicine should be to make the treatment more accessible to a wider swath of the population, including those who are poorer. Women of color have worse pregnancy outcomes than other groups. AI could help develop better treatment because it could help determine the reasons behind that disparity. There are some things that could be problematic though.
“There are concerns that people are all too fast to jump on new technology, and we really need to validate it,” Cedars said. “It needs to be something that's either cost neutral or decreases cost because we're already very expensive for the majority of people. And it needs to actually be proven and validated to improve outcome.”