Download This Paper
Preprints with The Lancet is a collaboration between The Lancet Group of journals and SSRN to facilitate the open sharing of preprints for early engagement, community comment, and collaboration. Preprints available here are not Lancet publications or necessarily under review with a Lancet journal. These preprints are early-stage research papers that have not been peer-reviewed. The usual SSRN checks and a Lancet-specific check for appropriateness and transparency have been applied. The findings should not be used for clinical or public health decision-making or presented without highlighting these facts. For more information, please see the FAQs.
IVFPred: A Novel Framework to Expediate Mining the Clinical Data of in Vitro Fertilization-Embryo Transfer
20 Pages Posted: 29 Jan 2024
More...Abstract
Machine learning (ML) models can be used to improve therapeutic schedule and predict outcomes for the women undergoing in vitro fertilization embryo transfer (IVF-ET). However, there is still lacking of a high-quality platform to facilitate the user-interactive modeling and the key predictors are under discussion. We aimed to build an analytical framework to predict reproductive outcomes of IVF-ET and ascertain the crucial clinical factors. Our study adopted various clinical features of 53,156 IVF-ET fresh cycles of 45,559 women from eight typical IVF-ET centers in China during 2012–2021. We built a framework to mine the IVF-ET data named “IVFPred”. It was cross-validated by six typical ML algorithms using data adopted. Overall, our developed freely-accessed platform for the public can provide an efficient streamlined framework to facilitate the related modeling by both online website operation (http://www.exposomex.cn/#/IVFPred) and R package (https://github.com/ExposomeX/IVFPred). Taking the clinical pregnancy as an application example, we found that very few parameters can realize the best prediction performance, e.g., female age (AGEF) and total number of high-quality embryos (TNHQE). In addition, better prediction model can be built by using the local center-based data than that using the multi-center. Furthermore, by using the partial dependence analysis, dramatic changes were found when AGEF ≥ 37 and NHQE ≤ 3. The anti-Müllerian hormone had weaker contribution of < 3% to prediction, and its testing may be less considered from the aspect of outcome prediction. Overall, our platform could potentially assist in building the ML models for improving the clinical practices of IVF-ET.
Funding: Dr. Bin Wang was supported by the National Key Research & Development Program of Ministry of Science and Technology of China (2022YFE0134900, 2023YFC3708305, and 2022YFC2704404), the National Natural Science Foundation of China (Grant No. 42077390, 41771527), and Yunnan Major Scientific and Technological Projects (Grant No. 202202AG050019).
Declaration of Interest: The authors declare no competing interests.
Ethical Approval: This study was approved by the Ethical Review Committee of Peking University People's Hospital (2021PHB358-001).
Keywords: assisted reproduction technology, machine learning, embryo quality, IVF-ET, analytical platform
Suggested Citation: Suggested Citation