Optimizing oocyte yield utilizing a machine learning model for dose and trigger decisions, a multi-center, prospective study

Abstract

Abstract The objective of this study was to evaluate clinical outcomes for patients undergoing IVF treatment where an artificial intelligence (AI) platform was utilized by clinicians to help determine the optimal starting dose of FSH and timing of trigger injection. This was a prospective clinical trial with historical control arm. Four physicians from two assisted reproductive technology treatment centers in the United States participated in the study. The treatment arm included patients undergoing autologous IVF cycles between December 2022–April 2023 where the physician use AI to help select starting dose of follicle stimulating hormone (FSH) and trigger injection timing (N = 291). The control arm included historical patients treated where the same doctor did not use AI between September 2021 and September 2022. The main outcome measures were total FSH used and average number of mature metaphase II (MII) oocytes. There was a non-significant trend towards improved patient outcomes and a reduction in FSH with physician use of AI. Overall, the average number of MIIs in the treatment vs. control arm was 12.20 vs 11.24 (improvement = 0.96, p = 0.16). The average number of oocytes retrieved in the treatment vs. control arm was 16.01 vs 14.54 (improvement = 1.47, p = 0.08). The average total FSH in the treatment arm was 3671.95 IUs and the average in the control arm was 3846.29 IUs (difference = −174.35 IUs, p = 0.13). These results suggests that AI can safely assist in refining the starting dose of FSH while narrowing down the timing of the trigger injection during ovarian stimulation, benefiting the patient in optimizing the count of MII oocytes retrieved.

Keywords

• Artificial intelligence
• Machine learning
• Embryo selection
• Clinical study