Discrete time survival analysis machine learning models for IVF prognosis across multiple cycles.
6 affiliations
- 1Department of Public Health and Primary Care, KU Leuven Campus Kulak Kortrijk, Kortrijk, Belgium.
- 2Department of Public Health and Primary Care, itec, imec Research Group at KU Leuven, Kortrijk, Belgium.
- 3Department of Public Health and Primary Care, KU Leuven, Leuven, Belgium.
- 4Department of Obstetrics and Gynecology, Leuven University Fertility Centre, University Hospital Leuven, Leuven, Belgium.
- 5Laboratory of Endometrium, Endometriosis & Reproductive Medicine, Department of Development and Regeneration, KU Leuven, Leuven, Belgium.
- 6Computer Science Department, State University of Londrina, Londrina, Brazil.
Abstract
Study question
Can discrete time survival analysis machine learning (ML) methods provide reliable and accurate predictions of cumulative live birth rate (CLBR) for patients both at diagnosis (pre-treatment) and after the first cycle of treatment (post-treatment)?Summary answer
We developed discrete time survival analysis ML models that provide personalized CLBR estimates of up to six complete IVF cycles with good discriminative performance, especially at the post-treatment phase.What is known already
The first relevant prognostic models to estimate IVF cumulative success rate across multiple cycles have moderate performance, with C-statistics ranging from 0.7 to 0.73 for both pre-treatment and post-treatment models.Study design size duration
This retrospective, single-center study included a cohort of 1429 couples who started their IVF treatment at a tertiary center between January 2014 and April 2021 as the development dataset. An additional 613 couples who started treatment at the same center between April 2021 and December 2023 were used as the temporal validation dataset.Participants/materials setting methods
Predictive performance was assessed using the following metrics: prediction error (Brier score, scaled Brier score), time-dependent AUC (cumulative/dynamic AUC, mean AUC, and time-dependent precision-recall AUC), and calibration (mean calibration, calibration plot, and calibration slope). Both pre-treatment and post-treatment models predicted the cumulative live birth success rates of up to six complete IVF cycles. Various discrete time survival analysis methods were trained, evaluated, and compared with the baseline models. Discrete time survival analysis models include the ones embedded with a generalized linear model (Logit) and tree-based ensembles (Bayesian Additive Regression Trees (BART), random forests). The baseline models are the refitted/recalibrated pre-treatment and post-treatment McLernon models, considering factors available in our data.Main results and the role of chance
The discrete time survival analysis models, especially BART, performed better on our data than the baseline models especially at the post-treatment phase. Validation results showed moderate discriminative ability for the pre-treatment model, with a mean AUC 0.69 (95% CI: 0.65, 0.73) and a scaled integrated Brier score (IBS) of 0.13 (13% performance improvement over Kaplan-Meier estimates). For the post-treatment model, discriminative power was strong, with a mean AUC of 0.82 (95% CI: 0.79, 0.85) and a scaled IBS of 0.19. In the BART model, the most important feature was female age and embryo utilization rate (EUR) in the pre- and post-treatment models, respectively.Large scale data
N/A.Limitations reasons for caution
Our study is limited by the retrospective and single-center design, including moderate sample size and long observation time span. Methodologically, several features were not present in our data when refitting the McLernon baseline models.Wider implications of the findings
This work illustrates how discrete time survival analysis ML models can be used to generate individualized, multi-cycle IVF prognoses. These ML models can be readily trained on local populations, which facilitates center-specific predictions. Our findings support ongoing efforts to improve patient counseling through data-driven communication about the individualized treatment plans.Funding
This study was supported by grants from the national Research Foundation-Flanders (FWO) (personal mandate 1235924N to F.K.N., research grant G046024N to C.V., T.T.S.) and the Flemish Government (Flanders AI Research program).Disclosures
The funders had no role in the study design, data collection or analysis, publication decision or manuscript preparation. There are no conflicts of interest to be declared.
Related
- Apparent diffusion coefficient histogram parameters of the whole tumor based on whole-body magnetic resonance imaging can better characterize the tumor burden of multiple myeloma and predict survival prognosis
- Predicting MCI-to-Alzheimer’s Conversion: A Systematic Benchmark of Classical, Machine Learning, and Deep Survival Models
- Development and temporal validation of nomograms and machine learning models for predicting overall and cancer-specific survival in primary distal cholangiocarcinoma: a SEER-based study
- Interpretable machine learning for predicting one-year prognosis in acute ischemic stroke.
- A Feasible Machine Learning Approach to Improve Cognitive Screening in Multiple Sclerosis
BibTeX
@article{shao2026discrete,
title = {{Discrete time survival analysis machine learning models for IVF prognosis across multiple cycles.}},
author = {TingTing Shao and Achilleas Ghinis and Eline Dancet and Johanna Devroe and Felipe Kenji Nakano and Celine Vens},
journal = {Human reproduction open},
year = {2026},
month = sep,
doi = {10.1093/hropen/hoag085},
url = {https://europepmc.org/article/MED/42921485},
note = {CC BY},
}
RIS
TY - JOUR
TI - Discrete time survival analysis machine learning models for IVF prognosis across multiple cycles.
AU - TingTing Shao
AU - Achilleas Ghinis
AU - Eline Dancet
AU - Johanna Devroe
AU - Felipe Kenji Nakano
AU - Celine Vens
PY - 2026
DA - 2026/09/23
JO - Human reproduction open
DO - 10.1093/hropen/hoag085
UR - https://europepmc.org/article/MED/42921485
AB - Study question Can discrete time survival analysis machine learning (ML) methods provide reliable and accurate predictions of cumulative live birth rate (CLBR) for patients both at diagnosis (pre-treatment) and after the first cycle of treatment (post-treatment)?Summary answer We developed discrete time survival analysis ML models that provide personalized CLBR estimates of up to six complete IVF cycles with good discriminative performance, especially at the post-treatment phase.What is known already The first relevant prognostic models to estimate IVF cumulative success rate across multiple cycles have moderate performance, with C-statistics ranging from 0.7 to 0.73 for both pre-treatment and post-treatment models.Study design size duration This retrospective, single-center study included a cohort of 1429 couples who started their IVF treatment at a tertiary center between January 2014 and April 2021 as the development dataset. An additional 613 couples who started treatment at the same center between April 2021 and December 2023 were used as the temporal validation dataset.Participants/materials setting methods Predictive performance was assessed using the following metrics: prediction error (Brier score, scaled Brier score), time-dependent AUC (cumulative/dynamic AUC, mean AUC, and time-dependent precision-recall AUC), and calibration (mean calibration, calibration plot, and calibration slope). Both pre-treatment and post-treatment models predicted the cumulative live birth success rates of up to six complete IVF cycles. Various discrete time survival analysis methods were trained, evaluated, and compared with the baseline models. Discrete time survival analysis models include the ones embedded with a generalized linear model (Logit) and tree-based ensembles (Bayesian Additive Regression Trees (BART), random forests). The baseline models are the refitted/recalibrated pre-treatment and post-treatment McLernon models, considering factors available in our data.Main results and the role of chance The discrete time survival analysis models, especially BART, performed better on our data than the baseline models especially at the post-treatment phase. Validation results showed moderate discriminative ability for the pre-treatment model, with a mean AUC 0.69 (95% CI: 0.65, 0.73) and a scaled integrated Brier score (IBS) of 0.13 (13% performance improvement over Kaplan-Meier estimates). For the post-treatment model, discriminative power was strong, with a mean AUC of 0.82 (95% CI: 0.79, 0.85) and a scaled IBS of 0.19. In the BART model, the most important feature was female age and embryo utilization rate (EUR) in the pre- and post-treatment models, respectively.Large scale data N/A.Limitations reasons for caution Our study is limited by the retrospective and single-center design, including moderate sample size and long observation time span. Methodologically, several features were not present in our data when refitting the McLernon baseline models.Wider implications of the findings This work illustrates how discrete time survival analysis ML models can be used to generate individualized, multi-cycle IVF prognoses. These ML models can be readily trained on local populations, which facilitates center-specific predictions. Our findings support ongoing efforts to improve patient counseling through data-driven communication about the individualized treatment plans.Funding This study was supported by grants from the national Research Foundation-Flanders (FWO) (personal mandate 1235924N to F.K.N., research grant G046024N to C.V., T.T.S.) and the Flemish Government (Flanders AI Research program).Disclosures The funders had no role in the study design, data collection or analysis, publication decision or manuscript preparation. There are no conflicts of interest to be declared.
SN - 2399-3529
ER -