AI might be able to boost the success rates of in vitro fertilization among hopeful families.

An AI program can help doctors analyze ultrasound scans to determine the best timing for a hormone injection associated with egg collection, researchers write in a new study published Jan. 8 in the journal Nature Communications.

Getting the timing right can optimize the number of eggs retrieved and improve the odds of having a successful pregnancy, researchers said.

“IVF provides help and hope for many patients who are unable to conceive but it’s an invasive, expensive, and time-consuming treatment,” co-senior researcher Dr. Ali Abbara, a clinician scientist at Imperial College London, said in a news release from the school.

“It can be heartbreaking when it fails, so it’s important to ensure that this treatment is as effective as possible,” Abbara continued. “AI can offer a new paradigm in how we deliver IVF treatment and could lead to better outcomes for patients.”

Doctors use ultrasound scans to monitor the size of follicles, which are small sacs in the ovaries containing eggs.

These scans determine when the woman will receive a “trigger” hormone injection to prepare the eggs for collection, researchers said.

The timing of the trigger is a key decision, as the injection is less effective if the follicles are too small or too large, researchers said.

For this study, researchers had an AI program analyze data on more than 19,000 patients who’d completed IVF treatment in one of 11 clinics across the UK and two in Poland between 2005 and 2023.

The AI determined that delivering the hormone injection when most follicles are between 13 mm to 18 mm improved the odds of retrieving more eggs and having more successful pregnancies, researchers said.

That’s more precise than the current method, in which doctors generally give the trigger when two or three follicles are larger than 17 mm to 18 mm.

“This is an exciting development as the findings suggest that we can use information from a much wider set of follicle sizes to decide when to give patients trigger shots rather than just the size of only the largest follicles - which is what is used in current clinical practice,” Abbara added.

The team plans to create an AI tool that will personalize IVF treatment and support doctor’s decision making at multiple steps along the IVF process.

“Where the stakes are so high for making the best possible decision, this technique can support doctors’ decision making and lead to better outcomes for patients,” co-senior author Thomas Heinis with Imperial College London’s Department of Computing said.

“Importantly, we expect computing power to improve exponentially in the near future, enabling us to make decisions using precise data in a way that hasn't been possible previously,” Heinis added.