Software application for predicting the health status of a child born with the use of assisted reproductive technologies, according to the mother's anamnesis

For many years, assisted reproductive technologies (ART) have been helping to conceive a child when this is not possible naturally. We can consider the ART protocol to be successful not only in case of pregnancy, but also in case of its successful completion: the birth of a healthy child. The article describes the creation of a software application for employees of ART centers, which helps to predict the outcome of the protocol, including the probability of pregnancy, the forecast of possible complications during its course, the forecast of the time and method of delivery, and the health group (1-5) of the born child. To create the application, we used data on 854 protocols implemented in 2016-2018, because of which 464 children were born. The analysis of their health contains information from birth to three years of age. The application uses sixteen binary classifiers, nine of which implement multi-class classifications of the term of delivery, the delivery method and children’s health groups. To implement multiclass inference, the “one-vs-all” strategy was used. Сross-validation was used to check the quality. The remaining 7 classifiers predict the likelihood of pregnancy and the occurrence of its complications: cervical incompetence, hypertensive disorders, placenta previa, gestational diabetes mellitus, violations of the amount of amniotic fluid and premature rupture of the membranes. We have built all the models based on the random forest algorithm using the Python programming language. The interface was created using the PyQT5 and QtDesigner libraries.

1. David J McLernon, Ewout W Steyerberg, Egbert R te Velde, Amanda J Lee, Siladitya Bhattacharya. Predicting the chances of a live birth after one or more complete cycles of in vitro fertilisation: population based study of linked cycle data from 113 873 women. BMJ. 2016;355(8082) DOI: 10.1136/bmj.i5735

2. Porcu, G., Lehert, P., Colella, C, Giorgetti C. Predicting live birth chances for women with multiple consecutive failing IVF cycles: a simple and accurate prediction for routine medical practice. Reproductive Biology and Endocrinology. 2013;11(1) DOI: 10.1186/1477-7827-11-1

3. Katarina Kebbon Vaegter, Tatevik Ghukasyan Lakic, Matts Olovsson, Lars Berglund, Thomas Brodin, Jan Holte. Which factors are most predictive for live birth after in vitro fertilization and intracytoplasmic sperm injection (IVF/ICSI) treatments? Analysis of 100 prospectively recorded variables in 8,400 IVF/ICSI single-embryo transfers. Fertility and Sterility. 2017;107(3):641-648.e2 DOI: 10.1016/j.fertnstert.2016.12.005.

4. Dhillon R.K., McLernon D.J., Smith P.P., Fishel S., Dowell K., Deeks J.J., Bhattacharya S., Coomarasamy A. Predicting the chance of live birth for women undergoing IVF: a novel pretreatment counselling tool. Hum Reprod. 2016;31(1):84-92 DOI: 10.1093/humrep/dev268

5. Bergh C., Wennerholm U.B. Long-term health of children conceived after assisted reproductive technology. Ups J Med Sci. 2020;125(2):152-157 DOI: 10.1080/0300973 4.2020.1729904

6. Heijligers M., Peeters A., van Montfoort A., Nijsten J., Janssen E., Gunnewiek F.K., de Rooy R., van Golde R., Coonen E., Meijer-Hoogeveen M., Broekmans F., van der Hoeven M., Arens Y., de Die-Smulders C. Growth, health, and motor development of 5-year-old children born after preimplantation genetic diagnosis. Fertil Steril. 2019;111(6):1151-1158 DOI: 10.1016/j.fertnstert.2019.01.035

7. Qiu J., Li P., Dong M., Xin X., Tan J. Personalized prediction of live birth prior to the first in vitro fertilization treatment: a machine learning method. J Transl Med. 2019;17:317 DOI: 10.1186/s12967-019-2062-5

8. Zare S., Thomsen M.R., Nayga R.M., Goudie A. Use of Machine Learning to Determine the Information Value of a BMI Screening Program. American Journal of Preventive Medicine. 2021;60(3):425-433. DOI: 10.1016/j.amepre.2020.10.016

9. Bose S., Kenyon C.C., Masino A.J. Personalized prediction of early childhood asthma persistence: A machine learning approach. PLOS ONE. 2021;16(3):1-17 DOI: 10.1371/journal.pone.0247784

10. Pedregosa et al. Scikit-learn: Machine Learning in Python. JMLR. 2011;12:2825-2830

11. Breiman L. Random Forests. Machine Learning. 2001;45:5-32

12. The RandomForestClassifier from the Scikit-Learn library. Available at: https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestClassifier.html (accessed 15.07.2021)

13. Morphological analyzer pymorphy2. Available at: https://pymorphy2.readthedocs.io/en /latest/ (accessed 15.07.2021)

14. The CountVectorizer from the Scikit-Learn library. Available at: https://scikit-learn.org/stable/modules/generated/sklearn.feature_extraction.text.CountVectorizer.html#sklearn.feature_extraction.text.CountVectorizer (accessed 15.07.2021)

15. The TfidfVectorizer from the Scikit-Learn library. Available at: https://scikit-learn.org/stable/modules/generated/sklearn.feature_extraction.text.TfidfVectorizer.html#sklearn.feature_extraction.text.TfidfVectorizer (accessed 15.07.2021)

16. Arthur D., Vassilvitskii S. K-means++: the advantages of careful seeding. Proc. of the Eighteenth Annu. ACM-SIAM Symp. on Discrete Algorithms, New Orleans, Louisiana, USA. 2007:1027-1035 DOI:10.1145/1283383.1283494

17. The K-Means clustering from the Scikit-Learn library. Available at: https://scikit-learn.org/stable/modules/generated/sklearn.cluster.KMeans.html?highlight=k%20means#sklearn.cluster.KMeans (accessed 15.07.2021)

18. Kursa M.B., Rudnicki W.R. Feature Selection with the Boruta Package. Journal of Statistical Software. 2010;36(11):1-13 DOI: 10.18637/jss.v036.i11

19. The Boruta algorithm implementation in Python. Available at: https://github.com/scikit-learncontrib/boruta_py (accessed 15.07.2021)

20. He, H.; Bai, Y.; Garcia, E.A.; Li, S. ADASYN: Adaptive synthetic sampling approach for imbalanced learning. Proceedings of the 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence), Hong Kong, China. 2008; 1322–1328. DOI: 10.1109/IJCNN.2008.4633969

21. Lemaitre G., Nogueira F., Aridas C.K. Imbalanced-learn: Imbalanced-learn: A python toolbox to tackle the curse of imbalanced datasets in machine learning. JMLR. 2017;18(17):1-5

22. Kohavi R. A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection. Proceedings of the Fourteenth International Joint Conference on Artificial Intelligence, San Mateo, CA. 1995;2(12):1137-1143

23. Stone M. Cross-Validatory Choice and Assessment of Statistical Predictions. Journal of the Royal Statistical Society, Series B (Methodological). 1974;36(2):111-147

24. Implementation of the StratifiedKFold cross-validation strategy in the Scikit-Learn library. Availiable at: https://scikit-learn.org/stable/modules/generated/sklearn.model _selection.StratifiedKFold.html (accessed 15.07.2021)

25. A set of Python libraries for creating a graphical interface based on the Qt5 platform. Available at: https://pypi.org/project/PyQt5/ (accessed 15.07.2021)

26. The Qt Designer. Available at: https://build-system.fman.io/qt-designer-download (accessed 15.07.2021)

27. The Pickle library. Available at: https://docs.python.org/3.6/library/pickle.html (accessed 15.07.2021)

28. Rifkin R., Klautau A. In Defense of One-Vs-All Classification. JMLR. 2004;5:101-104

29. Order of the Ministry of Health of the Russian Federation # 621 “O kompleksnoj ocenke sostoyaniya zdorov’ya detej” [About comprehensive assessment of the stage of health children] (in Russ). 2003