Research Progress on Clinical Prediction Models for Pregnancy Outcomes in Assisted Reproductive Technology

doi:10.12280/gjszjk.20240449

Abstract

Abstract:

The pregnancy outcome of assisted reproductive technology (ART) is influenced by a variety of factors, so the traditional assessment method based on clinical experience may be subjective and inaccurate. Clinical prediction models (CPM) can improve the accuracy and safety of treatment by comprehensively analyzing multimodal variable factors, contributing to precision medicine. Currently, a variety of pregnancy outcome CPMs have been constructed based on diverse algorithms, including not only traditional Logistic regression but also novel nonlinear machine learning algorithms such as random forests, neural networks and deep learning algorithms. Recent advancements in clinical research indicated that the CPMs constructed using a variety of algorithms have the high accuracy and practical potential in predicting pregnancy outcomes of ART. In future, the more accurate and reliable CPMs will be developed by collecting more diverse and representative clinical data, optimizing the model algorithms, engaging in multi-center collaborations and enhancing the generalization capabilities of the CPMs.

Key words: Reproductive techniques, assisted, Pregnancy outcome, Logistic regression, Machine learning, Clinical prediction model

WANG Cong, GONG Zheng, MA Sai-hua, HU Kai-yuan, LANG Meng-ran, XIA Tian. Research Progress on Clinical Prediction Models for Pregnancy Outcomes in Assisted Reproductive Technology[J]. Journal of International Reproductive Health/Family Planning, 2025, 44(1): 30-35.

Figures/Tables 1

References 36

[1]	Shacfe G, Turko R, Syed HH, et al. A DNA Methylation Perspective on Infertility[J]. Genes(Basel), 2023, 14(12):2132. doi: 10.3390/genes14122132.
[2]	Zegers-Hochschild F, Crosby JA, Musri C, et al. ART in Latin America: the Latin American Registry, 2020[J]. JBRA Assist Reprod, 2023, 27(3):514-538. doi: 10.5935/1518-0557.20230025.
[3]	Wong KY, Tan HH, Allen JC, et al. Outcomes and cost analysis of single-embryo transfer versus double-embryo transfer[J]. Womens Health(Lond), 2023,19:17455057231206312. doi: 10.1177/17455057231206312.
[4]	Davoodi Nik B, Hashemi Karoii D, Favaedi R, et al. Differential expression of ion channel coding genes in the endometrium of women experiencing recurrent implantation failures[J]. Sci Rep, 2024, 14(1):19822. doi: 10.1038/s41598-024-70778-9. pmid: 39192025
[5]	Cimadomo D, Craciunas L, Vermeulen N, et al. Definition, diagnostic and therapeutic options in recurrent implantation failure: an international survey of clinicians and embryologists[J]. Hum Reprod, 2021, 36(2):305-317. doi: 10.1093/humrep/deaa317.
[6]	Bala R, Singh V, Rajender S, et al. Environment, Lifestyle, and Female Infertility[J]. Reprod Sci, 2021, 28(3):617-638. doi: 10.1007/s43032-020-00279-3.
[7]	Li YH, Li HR, Wang PH. Parameters to predict the pregnancy in assisted reproductive technology[J]. J Chin Med Assoc, 2019, 82(4):249-250. doi: 10.1097/JCMA.0000000000000060.
[8]	Methorst C, Perrin J, Faix A, et al. Male infertility, environment and lifestyle[J]. Prog Urol, 2023, 33(13):613-623. doi: 10.1016/j.purol.2023.09.014. pmid: 38012907
[9]	van Smeden M, Reitsma JB, Riley RD, et al. Clinical prediction models: diagnosis versus prognosis[J]. J Clin Epidemiol, 2021, 132:142-145. doi: 10.1016/j.jclinepi.2021.01.009. pmid: 33775387
[10]	Strandberg R, Jepsen P, Hagström H. Developing and validating clinical prediction models in hepatology-An overview for clinicians[J]. J Hepatol, 2024, 81(1):149-162. doi: 10.1016/j.jhep.2024.03.030.
[11]	刘志强, 熊风, 张宏展, 等. IVF-ET妊娠结局预测模型的研究进展[J]. 生殖医学杂志, 2021, 30(5):695-700. doi: 10.3969/j.issn.1004-3845.2021.05.025.
[12]	Shingshetty L, Cameron NJ, Mclernon DJ, et al. Predictors of success after in vitro fertilization[J]. Fertil Steril, 2024, 121(5):742-751. doi: 10.1016/j.fertnstert.2024.03.003. pmid: 38492930
[13]	于医萍, 高一博, 方兰兰, 等. 机器学习在体外受精-胚胎移植技术中的应用[J]. 中华生殖与避孕杂志, 2021, 41(10):883-892. doi: 10.3760/cma.j.cn101441-20200428-00251.
[14]	Jiang S, Li L, Li F, et al. Establishment of predictive model for analyzing clinical pregnancy outcome based on IVF-ET and ICSI assisted reproductive technology[J]. Saudi J Biol Sci, 2020, 27(4):1049-1056. doi: 10.1016/j.sjbs.2020.02.021. pmid: 32256165
[15]	Zhu S, Jiang W, Sun Y, et al. Nomogram to predict the probability of clinical pregnancy in women with poor ovarian response undergoing in vitro fertilization/ intracytoplasmic sperm injection cycles[J]. Arch Gynecol Obstet, 2024, 310(3):1697-1707. doi: 10.1007/s00404-024-07598-9.
[16]	Balachandren N, Salman M, Diu NL, et al. Ovarian reserve as a predictor of cumulative live birth[J]. Eur J Obstet Gynecol Reprod Biol, 2020, 252:273-277. doi: 10.1016/j.ejogrb.2020.06.063. pmid: 32645642
[17]	Sun X, Yao F, Yin C, et al. Independent value of PMOI on hCG day in predicting pregnancy outcomes in IVF/ICSI cycles[J]. Front Endocrinol(Lausanne), 2023,14:1086998. doi: 10.3389/fendo.2023.1086998.
[18]	Meng S, Shi C, Jia Y, et al. A combined clinical and specific genes′ model to predict live birth for in vitro fertilization and embryo transfer patients[J]. BMC Pregnancy Childbirth, 2023, 23(1):702. doi: 10.1186/s12884-023-05988-6.
[19]	Chen L, Jiang R, Jiang Y, et al. A validated model for individualized prediction of pregnancy outcome in woman after fresh cycle of Day 5 single blastocyst transfer[J]. Sci Rep, 2023, 13(1):10016. doi: 10.1038/s41598-023-36824-8. pmid: 37340007
[20]	Raef B, Ferdousi R. A Review of Machine Learning Approaches in Assisted Reproductive Technologies[J]. Acta Inform Med, 2019, 27(3):205-211. doi: 10.5455/aim.2019.27.205-211.
[21]	Li L, Cui X, Yang J, et al. Using feature optimization and LightGBM algorithm to predict the clinical pregnancy outcomes after in vitro fertilization[J]. Front Endocrinol(Lausanne), 2023, 14:1305473. doi: 10.3389/fendo.2023.1305473.
[22]	Yang L, Peavey M, Kaskar K, et al. Development of a dynamic machine learning algorithm to predict clinical pregnancy and live birth rate with embryo morphokinetics[J]. F S Rep, 2022, 3(2):116-123. doi: 10.1016/j.xfre.2022.04.004.
[23]	Chen F, Chen Y, Mai Q. Multi-Omics Analysis and Machine Learning Prediction Model for Pregnancy Outcomes After Intracytoplasmic Sperm Injection-in vitro Fertilization[J]. Front Public Health, 2022,10:924539. doi: 10.3389/fpubh.2022.924539.
[24]	Hu J, Szymczak S. A review on longitudinal data analysis with random forest[J]. Brief Bioinform, 2023, 24(2):bbad002. doi: 10.1093/bib/bbad002.
[25]	Liang R, An J, Zheng Y, et al. Predicting and improving the probability of live birth for women undergoing frozen-thawed embryo transfer: a data-driven estimation and simulation model[J]. Comput Methods Programs Biomed, 2021,198:105780. doi: 10.1016/j.cmpb.2020.105780.
[26]	Wang CW, Kuo CY, Chen CH, et al. Predicting clinical pregnancy using clinical features and machine learning algorithms in in vitro fertilization[J]. PLoS One, 2022, 17(6):e0267554. doi: 10.1371/journal.pone.0267554.
[27]	Yang H, Liu F, Ma Y, et al. Clinical pregnancy outcomes prediction in vitro fertilization women based on random forest prediction model: A nested case-control study[J]. Medicine(Baltimore), 2022, 101(49):e32232. doi: 10.1097/MD.0000000000032232.
[28]	Yang R, Yu Y. Artificial Convolutional Neural Network in Object Detection and Semantic Segmentation for Medical Imaging Analysis[J]. Front Oncol, 2021,11:638182. doi: 10.3389/fonc.2021.638182.
[29]	Liu L, Liang H, Yang J, et al. Clinical data-based modeling of IVF live birth outcome and its application[J]. Reprod Biol Endocrinol, 2024, 22(1):76. doi: 10.1186/s12958-024-01253-3.
[30]	Enatsu N, Miyatsuka I, An LM, et al. A novel system based on artificial intelligence for predicting blastocyst viability and visualizing the explanation[J]. Reprod Med Biol, 2022, 21(1):e12443. doi: 10.1002/rmb2.12443.
[31]	Liu H, Zhang Z, Gu Y, et al. Development and evaluation of a live birth prediction model for evaluating human blastocysts from a retrospective study[J]. Elife, 2023,12:e83662. doi: 10.7554/eLife.83662.
[32]	Thirunavukarasu R, C GPD, R G, et al. Towards computational solutions for precision medicine based big data healthcare system using deep learning models: A review[J]. Comput Biol Med, 2022, 149:106020. doi: 10.1016/j.compbiomed.2022.106020.
[33]	Ueno S, Berntsen J, Ito M, et al. Pregnancy prediction performance of an annotation-free embryo scoring system on the basis of deep learning after single vitrified-warmed blastocyst transfer: a single-center large cohort retrospective study[J]. Fertil Steril, 2021, 116(4):1172-1180. doi: 10.1016/j.fertnstert.2021.06.001. pmid: 34246469
[34]	Mapstone C, Hunter H, Brison D, et al. Deep learning pipeline reveals key moments in human embryonic development predictive of live birth after in vitro fertilization[J]. Biol Methods Protoc, 2024, 9(1):bpae052. doi: 10.1093/biomethods/bpae052.
[35]	Huang B, Zheng S, Ma B, et al. Using deep learning to predict the outcome of live birth from more than 10,000 embryo data[J]. BMC Pregnancy Childbirth, 2022, 22(1):36. doi: 10.1186/s12884-021-04373-5.
[36]	Ratna MB, Bhattacharya S, Abdulrahim B, et al. A systematic review of the quality of clinical prediction models in in vitro fertilisation[J]. Hum Reprod, 2020, 35(1):100-116. doi: 10.1093/humrep/dez258.

文献	发表年份（年）	研究类型	样本量	纳入影响因素	算法	结局	模型效果（AUC）
Balachandren等^[16]	2020	回顾性单中心病例对照研究	516个周期	年龄、AMH、FSH	逻辑回归	活产	0.680
Liang等^[25]	2021	回顾性单中心病例对照研究	2 189个周期	年龄、BMI、HOMA-IR、FSH、LH、E₂、子宫内膜厚度、胚胎总数、ET次数	随机森林	活产	0.830
Ueno等^[33]	2021	回顾性单中心病例对照研究	3 018个周期	年龄、胚胎评分	深度学习	临床妊娠	>0.600
Yang等^[22]	2022	回顾性单中心病例对照研究	367个胚胎	胚胎形态动力学	随机森林	临床妊娠	0.910
Chen等^[23]	2022	回顾性单中心病例对照研究	338个位点	卵丘细胞甲基化特征	随机森林	临床妊娠	0.880
Wang等^[26]	2022	回顾性单中心病例对照研究	17 288例患者	年龄、不孕原因、不孕年限、卵巢刺激方案、冷冻胚胎总数	随机森林	临床妊娠	0.721
Yang等^[27]	2022	回顾性单中心病例对照研究	369例患者	年龄、BMI、周期数、红细胞压积、LH、孕酮、FSH和子宫内膜厚度	随机森林	临床妊娠	0.817
Enatsu等^[30]	2022	回顾性单中心病例对照研究	19 342个囊胚	囊胚图像、胚胎数据	神经网络	临床妊娠	0.710
Huang等^[35]	2022	回顾性单中心病例对照研究	33 738个胚胎	胚胎数据	深度学习	活产	0.968
Sun等^[17]	2023	回顾性单中心病例对照研究	1 239个周期	PMOI	逻辑回归	临床妊娠	0.621
Meng等^[18]	2023	前瞻性单中心病例对照研究	39例患者	排卵功能障碍、GAST、GPX3、THBS2	逻辑回归	活产	0.842
Chen等^[19]	2023	回顾性单中心病例对照研究	86例患者	AMH、E₂、年龄、精子DFI、hsa-miR-199a-3p、hsa-miR-199a-5p、hsa-miR-99a-5p	逻辑回归	临床妊娠	0.853
Li等^[21]	2023	回顾性双中心病例对照研究	840例患者	不孕年限、BMI、扳机日E₂、扳机日子宫内膜厚度	LightGBM机器学习	临床妊娠	0.904
Liu等^[31]	2023	回顾性单中心病例对照研究	17 580个囊胚	囊胚图像、不孕夫妇的临床特征	神经网络	活产	0.770
Zhu等^[15]	2024	回顾性单中心病例对照研究	969例患者	年龄、BMI、AFC、AMH、成熟卵母细胞数、移植胚胎数、移植胚胎质量	逻辑回归	临床妊娠	0.752
Liu等^[29]	2024	回顾性单中心病例对照研究	1 405个周期	年龄、卵巢敏感指数、控制性卵巢刺激方案、Gn起始剂量、扳机日子宫内膜厚度、扳机日孕酮水平、ET方案	神经网络	活产	0.726
Mapstone等^[34]	2024	回顾性单中心病例对照研究	700个胚胎	胚胎的延时视频图像数据	深度学习	活产	0.680

文献	发表年份（年）	研究类型	样本量	纳入影响因素	算法	结局	模型效果（AUC）
Balachandren等^[16]	2020	回顾性单中心病例对照研究	516个周期	年龄、AMH、FSH	逻辑回归	活产	0.680
Liang等^[25]	2021	回顾性单中心病例对照研究	2 189个周期	年龄、BMI、HOMA-IR、FSH、LH、E₂、子宫内膜厚度、胚胎总数、ET次数	随机森林	活产	0.830
Ueno等^[33]	2021	回顾性单中心病例对照研究	3 018个周期	年龄、胚胎评分	深度学习	临床妊娠	>0.600
Yang等^[22]	2022	回顾性单中心病例对照研究	367个胚胎	胚胎形态动力学	随机森林	临床妊娠	0.910
Chen等^[23]	2022	回顾性单中心病例对照研究	338个位点	卵丘细胞甲基化特征	随机森林	临床妊娠	0.880
Wang等^[26]	2022	回顾性单中心病例对照研究	17 288例患者	年龄、不孕原因、不孕年限、卵巢刺激方案、冷冻胚胎总数	随机森林	临床妊娠	0.721
Yang等^[27]	2022	回顾性单中心病例对照研究	369例患者	年龄、BMI、周期数、红细胞压积、LH、孕酮、FSH和子宫内膜厚度	随机森林	临床妊娠	0.817
Enatsu等^[30]	2022	回顾性单中心病例对照研究	19 342个囊胚	囊胚图像、胚胎数据	神经网络	临床妊娠	0.710
Huang等^[35]	2022	回顾性单中心病例对照研究	33 738个胚胎	胚胎数据	深度学习	活产	0.968
Sun等^[17]	2023	回顾性单中心病例对照研究	1 239个周期	PMOI	逻辑回归	临床妊娠	0.621
Meng等^[18]	2023	前瞻性单中心病例对照研究	39例患者	排卵功能障碍、GAST、GPX3、THBS2	逻辑回归	活产	0.842
Chen等^[19]	2023	回顾性单中心病例对照研究	86例患者	AMH、E₂、年龄、精子DFI、hsa-miR-199a-3p、hsa-miR-199a-5p、hsa-miR-99a-5p	逻辑回归	临床妊娠	0.853
Li等^[21]	2023	回顾性双中心病例对照研究	840例患者	不孕年限、BMI、扳机日E₂、扳机日子宫内膜厚度	LightGBM机器学习	临床妊娠	0.904
Liu等^[31]	2023	回顾性单中心病例对照研究	17 580个囊胚	囊胚图像、不孕夫妇的临床特征	神经网络	活产	0.770
Zhu等^[15]	2024	回顾性单中心病例对照研究	969例患者	年龄、BMI、AFC、AMH、成熟卵母细胞数、移植胚胎数、移植胚胎质量	逻辑回归	临床妊娠	0.752
Liu等^[29]	2024	回顾性单中心病例对照研究	1 405个周期	年龄、卵巢敏感指数、控制性卵巢刺激方案、Gn起始剂量、扳机日子宫内膜厚度、扳机日孕酮水平、ET方案	神经网络	活产	0.726
Mapstone等^[34]	2024	回顾性单中心病例对照研究	700个胚胎	胚胎的延时视频图像数据	深度学习	活产	0.680