Abstract
Spontaneous preterm birth prediction from transvaginal ultrasound images is a challenging task of profound interest in gynecological obstetrics. Existing works are often validated on small datasets and may lack validation of model calibration and interpretation. In this paper, we present a comprehensive study of methods for predicting preterm birth from transvaginal ultrasound using a large clinical dataset. We propose a shape- and spatially-aware network that leverages segmentation predictions and pixel spacing information as additional input to enhance predictions. Our model demonstrates competitive performance on our benchmark, providing additional interpretation and achieving the highest performance across both clinical and machine learning baselines. Through our evaluation, we provide additional insights which we hope may lead to more accurate predictions of preterm births going forwards.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Ahmed, W.S., Madny, E., Habash, Y., Ibrahim, Z., Morsy, A., Said, M.: Ultrasonographic wall thickness measurement of the upper and lower uterine segments in the prediction of the progress of preterm labour. Clin. Exp. Obstetr. Gynecol. 42(3), 331–335 (2015)
Akazawa, M., Hashimoto, K.: Prediction of preterm birth using artificial intelligence: a systematic review. J. Obstetr. Gynaecol. 42(6), 1662–1668 (2022)
Baños, N., et al.: Quantitative analysis of cervical texture by ultrasound in mid-pregnancy and association with spontaneous preterm birth. Ultrasound Obstetr. Gynecol. 51(5), 637–643 (2018)
Baumgartner, C.F., et al.: SonoNet: real-time detection and localisation of fetal standard scan planes in freehand ultrasound. IEEE TMI 36(11) (2017)
Burgos-Artizzu, X.P., et al.: Mid-trimester prediction of spontaneous preterm birth with automated cervical quantitative ultrasound texture analysis and cervical length: a prospective study. Sci. Rep. 11(1), 1–7 (2021)
Bustamante, D., et al.: Cervix ultrasound texture analysis to differentiate between term and preterm birth pregnancy: a machine learning approach. In: IEEE IUS, pp. 1–4. IEEE IUS (2022)
Coutinho, C., et al.: Isuog practice guidelines: role of ultrasound in the prediction of spontaneous preterm birth. Ultrasound Obstetr. Gynecol. Off. J. Int. Soc. Ultrasound Obstetr. Gynecol. 60(3), 435–456 (2022)
Dagle, A.B., et al.: Automated segmentation of cervical anatomy to interrogate preterm birth. In: Licandro, R., Melbourne, A., Abaci Turk, E., Macgowan, C., Hutter, J. (eds.) PIPPI 2022. LNCS, vol. 13575, pp. 48–59. Springer, Cham (2022). https://doi.org/10.1007/978-3-031-17117-8_5
Dziadosz, M., et al.: Uterocervical angle: a novel ultrasound screening tool to predict spontaneous preterm birth. Am. J. Obstetr. Gynecol. 215(3) (2016)
Farràs Llobet, A., et al.: The uterocervical angle and its relationship with preterm birth. J. Maternal-Fetal Neonatal Med. 31(14), 1881–1884 (2018)
Fiset, S., Martel, A., Glanc, P., Barrett, J., Melamed, N.: Prediction of spontaneous preterm birth among twin gestations using machine learning and texture analysis of cervical ultrasound images. U Toronto Med. J. 96(1) (2019)
Giakoumoglou, N.: Pyfeats (2021). https://doi.org/10.5281/zenodo.6783286
Hemming, V.G., Overall, J.C., Jr., Britt, M.R.: Nosocomial infections in a newborn intensive-care unit: results of forty-one months of surveillance. N. Engl. J. Med. 294(24), 1310–1316 (1976)
Hermann, K., Chen, T., Kornblith, S.: The origins and prevalence of texture bias in convolutional neural networks. In: NeurIPS, vol. 33, pp. 19000–19015 (2020)
Lin, M., Feragen, A., Bashir, Z., Tolsgaard, M.G., Christensen, A.N.: I saw, I conceived, I concluded: progressive concepts as bottlenecks. arXiv:2211.10630 (2022)
Lin, M., et al.: DTU-net: learning topological similarity for curvilinear structure segmentation. In: Frangi, A., de Bruijne, M., Wassermann, D., Navab, N. (eds.) IPMI 2023. LNCS, vol. 13939, pp. 654–666. Springer, Cham (2023). https://doi.org/10.1007/978-3-031-34048-2_50
Luechathananon, S., Songthamwat, M., Chaiyarach, S.: Uterocervical angle and cervical length as a tool to predict preterm birth in threatened preterm labor. Int. J. Women’s Health, 153–159 (2021)
Mikolaj, K., et al.: Removing confounding information from fetal ultrasound images. arXiv:2303.13918 (2023)
Petersen, E., Ganz, M., Holm, S., Feragen, A.: On (assessing) the fairness of risk score models. In: FAccT, pp. 817–829 (2023)
Pizzella, S., El Helou, N., Chubiz, J., Wang, L.V., Tuuli, M.G., England, S.K., Stout, M.J.: Evolving cervical imaging technologies to predict preterm birth. Seminars Immunopathol. 42, 385–396 (2020). Springer
Sepúlveda-Martínez, A., Diaz, F., Muñoz, H., Valdés, E., Parra-Cordero, M.: Second-trimester anterior cervical angle in a low-risk population as a marker for spontaneous preterm delivery. Fetal Diagn. Ther. 41(3), 220–225 (2017)
Telea, A.: An image inpainting technique based on the fast marching method. J. Graph. Tools 9(1), 23–34 (2004)
Ven, V.D., et al.: The capacity of mid-pregnancy cervical length to predict preterm birth in low-risk women: a national cohort study. Acta Obstet. Gynecol. Scand. 94(11), 1223–1234 (2015)
Vogel, J.P., Chawanpaiboon, S., Moller, A.B., Watananirun, K., Bonet, M., Lumbiganon, P.: The global epidemiology of preterm birth. Best Pract. Res. Clin. Obstetr. Gynaecol. 52, 3–12 (2018)
Włodarczyk, T., et al.: Spontaneous preterm birth prediction using convolutional neural networks. In: Hu, Y., et al. (eds.) ASMUS/PIPPI -2020. LNCS, vol. 12437, pp. 274–283. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-60334-2_27
Włodarczyk, T., et al.: Machine learning methods for preterm birth prediction: a review. Electronics 10(5), 586 (2021)
Włodarczyk, T., et al.: Estimation of preterm birth markers with U-net segmentation network. In: Wang, Q., et al. (eds.) PIPPI/SUSI -2019. LNCS, vol. 11798, pp. 95–103. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-32875-7_11
Yang, Q., et al.: Reporting and risk of bias of prediction models based on machine learning methods in preterm birth: a systematic review. Acta Obstet. Gynecol. Scand. 102(1), 7–14 (2023)
Zuo, J., McFarlin, B.L., Simpson, D.G., O’Brien, W.D., Han, A.: Automated region of interest placement on cervical ultrasound images for assessing preterm birth risk. J. Acoust. Soc. Am. 153(3), A352–A352 (2023)
Acknowledgments
This work was supported by the Pioneer Centre for AI (DNRF grant nr P1), the DIREC project EXPLAIN-ME (9142-00001B), and the Novo Nordisk Foundation through the Center for Basic Machine Learning Research in Life Science (NNF20OC0062606).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Pegios, P. et al. (2023). Leveraging Shape and Spatial Information for Spontaneous Preterm Birth Prediction. In: Kainz, B., Noble, A., Schnabel, J., Khanal, B., Müller, J.P., Day, T. (eds) Simplifying Medical Ultrasound. ASMUS 2023. Lecture Notes in Computer Science, vol 14337. Springer, Cham. https://doi.org/10.1007/978-3-031-44521-7_6
Download citation
DOI: https://doi.org/10.1007/978-3-031-44521-7_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-44520-0
Online ISBN: 978-3-031-44521-7
eBook Packages: Computer ScienceComputer Science (R0)
