Prediction of Placenta Accreta Spectrum by Machine Learning Methods and Determination of Candidate Biomarkers

Abstract

Placenta accreta spectrum (PAS) disorders; Abnormal adhesion of placental villi to the myometrium associated with endometrial trauma or dysplasia. Placenta previa and previous cesarean section operations are two major risk factors for PAS disorders. It is usually diagnosed by ultrasound examinations performed during pregnancy follow-up. After this diagnosis is made, a very careful and strict pregnancy follow-up should be done. If the diagnosis is made during pregnancy, the delivery should be done by cesarean section and the bleeding that the mother will experience should be stopped with an appropriate method. However, no protein candidate to be used in clinical diagnosis has been found so far. The aim of this study is to identify candidate biomarkers that can be used in the diagnosis and follow-up of PAS with machine learning methods. In this study, proteomic data obtained from 26 women with and without PAS were used. After using the Lasso method as the variable selection method, machine learning models (XGBoost, Adaboost) were created with 5-fold cross-validation. Accuracy, Balanced accuracy, Sensitivity, Specificity, Positive Predictive Value, Negative Predictive Value, F1-Score, MCC and G-mean metrics were used in the performance evaluation of the models created. When the performance metrics of the two models are compared, the best result belongs to the XGBoost machine learning model. Therefore, the Accuracy, Balanced accuracy, Sensitivity, Specificity, Positive Predictive Value, Negative Predictive Value, F1-Score, MCC, and G-mean performance criteria for the XGBoost model are 0.962, 0.950, 1.00, 0.90, 0.94, 1.00, 0.97, 0.92, and 0.97, respectively. As a result, considering the experimental results, it can be said that the created machine learning model is quite successful in classifying PAS. In addition, it can be said that KDR and AMH proteins are candidate biomarkers that can be used in the diagnosis and follow-up of PAS according to the significance of the variables related to the model.

Keywords

• Placenta Accreta Spectrum
• Biomarker
• Machine learning
• Machine learning
• XGBoost
• Proteomics

References

1. Marshall, N.E., Fu R., Guise, J.M.(2011). Impact of multiple cesarean deliveries on maternal morbidity: a systematic review. American journal of obstetrics and gynecology. 205(3):262. e1-. e8.
2. Miller, R., Timor-Tritsch, I.E., Gyamfi-Bannerman, C., Medicine SfM-F. (2020) Society for Maternal-Fetal Medicine (SMFM) consult series# 49: cesarean scar pregnancy. American Journal of Obstetrics and Gynecology.222(5):B2-B14.
3. Bailit, J.L., Grobman, W., Rice, M.M., Reddy, U.M., Wapner, R.J., Varner, M.W., et al. (2015) Morbidly adherent placenta treatments and outcomes. Obstetrics and gynecology.125(3):683.
4. Fitzpatrick, K., Sellers, S., Spark, P., Kurinczuk, J., Brocklehurst, P., Knight, M. (2015) The Management and Outcomes of Placenta Accreta, Increta, and Percreta in the United Kingdom: A Population-based Descriptive Study. Obstetric Anesthesia Digest. 35(1):24-5.
5. Greener, J.G., Kandathil, S.M., Moffat, L., Jones, D.T. (2022) A guide to machine learning for biologists. Nature Reviews Molecular Cell Biology. 23(1):40-55.
6. Shainker, S.A., Silver, R.M., Modest, A.M., Hacker, M.R., Hecht, J.L., Salahuddin, S., et al. (2020) Placenta accreta spectrum: biomarker discovery using plasma proteomics. American Journal of Obstetrics and Gynecology. 223(3):433. e1-. e14.
7. Chen, T., He, T., Benesty, M., Khotilovich, V., Tang, Y., Cho, H., et al. (2015) Xgboost: extreme gradient boosting. R package version 04-2. 1(4):1-4.
8. Schapire, R.E. (2013) Explaining adaboost. Empirical inference: Springer. p. 37-52.

9. Ranstam, J., Cook, J. (2018) LASSO regression. Journal of British Surgery. 105(10):1348.
10. Silver, R.M., Branch, D.W. (2018) Placenta accreta spectrum. New England Journal of Medicine. 378(16):1529-36.
11. Jauniaux, E., Alfirevic, Z., Bhide, A., Belfort, M., Burton, G., Collins, S., et al. (2018) Placenta Praevia and Placenta Accreta: Diagnosis and Management: Green-top Guideline No. 27a. Bjog. 126(1):e1-e48.
12. Berkley, E.M., Abuhamad, A.Z. (2013) Prenatal diagnosis of placenta accreta: is sonography all we need? Journal of ultrasound in medicine. 32(8):1345-50.
13. D'antonio, F., Iacovella, C., Palacios‐Jaraquemada, J., Bruno, C., Manzoli, L., Bhide, A. (2014) Prenatal identification of invasive placentation using magnetic resonance imaging: systematic review and meta‐analysis. Ultrasound in Obstetrics & Gynecology. 44(1):8-16.
14. Riteau, A-S., Tassin, M., Chambon, G., Le, Vaillant. C., de Laveaucoupet, J., Quere, M-P., et al. (2014) Accuracy of ultrasonography and magnetic resonance imaging in the diagnosis of placenta accreta. PLoS One. 9(4):e94866.
15. Balcacer, P., Pahade, J., Spektor, M., Staib, L., Copel, J.A., McCarthy, S. (2016) Magnetic resonance imaging and sonography in the diagnosis of placental invasion. Journal of Ultrasound in Medicine. 35(7):1445-56.
16. Shazly, S.A., Hortu, I., Shih, J-C., Melekoglu, R., Fan, S., Ahmed, FuA., et al. (2021) Prediction of clinical outcomes in women with placenta accreta spectrum using machine learning models: an international multicenter study. The Journal of Maternal-Fetal & Neonatal Medicine. 1-10.
17. Romeo, V., Ricciardi, C., Cuocolo, R., Stanzione, A., Verde, F., Sarno, L,, et al. (2019) Machine learning analysis of MRI-derived texture features to predict placenta accreta spectrum in patients with placenta previa. Magnetic resonance imaging. 64:71-6.