An interpretable machine learning model for individualized gonadotrophin starting dose selection during ovarian stimulation.

RESEARCH QUESTION: Can we develop an interpretable machine learning model that optimizes starting gonadotrophin dose selection in terms of mature oocytes (metaphase II [MII]), fertilized oocytes (2 pronuclear [2PN]) and usable blastocysts? DESIGN: This was a retrospective study of patients undergoing autologous IVF cycles from 2014 to 2020 (n = 18,591) in three assisted reproductive technology centres in the USA. For each patient cycle, an individual dose-response curve was generated from the 100 most similar patients identified using a K-nearest neighbours model. Patients were labelled as dose-responsive if their dose-response curve showed a region that maximized MII oocytes, and flat-responsive otherwise. RESULTS: Analysis of the dose-response curves showed that 30% of cycles were dose-responsive and 64% were flat-responsive. After propensity score matching, patients in the dose-responsive group who received an optimal starting dose of FSH had on average 1.5 more MII oocytes, 1.2 more 2PN embryos and 0.6 more usable blastocysts using 10 IU less of starting FSH and 195 IU less of total FSH compared with patients given non-optimal doses. In the flat-responsive group, patients who received a low starting dose of FSH had on average 0.3 more MII oocytes, 0.3 more 2PN embryos and 0.2 more usable blastocysts using 149 IU less of starting FSH and 1375 IU less of total FSH compared with patients with a high starting dose. CONCLUSIONS: This study demonstrates retrospectively that using a machine learning model for selecting starting FSH can achieve optimal laboratory outcomes while reducing the amount of starting and total FSH used.

An interpretable machine learning model for predicting the optimal day of trigger during ovarian stimulation.

OBJECTIVE: To develop an interpretable machine learning model for optimizing the day of trigger in terms of mature oocytes (MII), fertilized oocytes (2PNs), and usable blastocysts. DESIGN: Retrospective study. SETTING: A group of three assisted reproductive technology centers in the United States. PATIENT(S): Patients undergoing autologous in vitro fertilization cycles from 2014 to 2020 (n = 30,278). INTERVENTION(S): None. MAIN OUTCOME MEASURE(S): Average number of MII oocytes, 2PNs, and usable blastocysts. RESULT(S): A set of interpretable machine learning models were developed using linear regression with follicle counts and estradiol levels. When using the model to make day-by-day predictions of trigger or continuing stimulation, possible early and late triggers were identified in 48.7% and 13.8% of cycles, respectively. After propensity score matching, patients with early triggers had on average 2.3 fewer MII oocytes, 1.8 fewer 2PNs, and 1.0 fewer usable blastocysts compared with matched patients with on-time triggers, and patients with late triggers had on average 2.7 fewer MII oocytes, 2.0 fewer 2PNs, and 0.7 fewer usable blastocysts compared with matched patients with on-time triggers. CONCLUSION(S): This study demonstrates that it is possible to develop an interpretable machine learning model for optimizing the day of trigger. Using our model has the potential to improve outcomes for many in vitro fertilization patients.

Characterization of an artificial intelligence model for ranking static images of blastocyst stage embryos.

OBJECTIVE: To perform a series of analyses characterizing an artificial intelligence (AI) model for ranking blastocyst-stage embryos. The primary objective was to evaluate the benefit of the model for predicting clinical pregnancy, whereas the secondary objective was to identify limitations that may impact clinical use. DESIGN: Retrospective study. SETTING: Consortium of 11 assisted reproductive technology centers in the United States. PATIENT(S): Static images of 5,923 transferred blastocysts and 2,614 nontransferred aneuploid blastocysts. INTERVENTION(S): None. MAIN OUTCOME MEASURE(S): Prediction of clinical pregnancy (fetal heartbeat). RESULT(S): The area under the curve of the AI model ranged from 0.6 to 0.7 and outperformed manual morphology grading overall and on a per-site basis. A bootstrapped study predicted improved pregnancy rates between +5% and +12% per site using AI compared with manual grading using an inverted microscope. One site that used a low-magnification stereo zoom microscope did not show predicted improvement with the AI. Visualization techniques and attribution algorithms revealed that the features learned by the AI model largely overlap with the features of manual grading systems. Two sources of bias relating to the type of microscope and presence of embryo holding micropipettes were identified and mitigated. The analysis of AI scores in relation to pregnancy rates showed that score differences of ≥0.1 (10%) correspond with improved pregnancy rates, whereas score differences of <0.1 may not be clinically meaningful. CONCLUSION(S): This study demonstrates the potential of AI for ranking blastocyst stage embryos and highlights potential limitations related to image quality, bias, and granularity of scores.