Computational Approaches for Connecting Maternal Stress to Preterm Birth - 03/05/24

Doi : 10.1016/j.clp.2024.02.003

Amin Mirzaei, MS ^a, Bjarne C. Hiller, MS ^a, Ina A. Stelzer, PhD ^b, Kristin Thiele, PhD ^d, Yuqi Tan, PhD ^c, Martin Becker, PhD ^a,^⁎
^a Department of Computer Science and Electrical Engineering, Institute for Visual and Analytic Computing, Universität Rostock, Albert-Einstein-Straße 22, 18059 Rostock, Germany
^b Department of Pathology, University of California San Diego, GPL/CMM-West, 9500 Gilman Drive, La Jolla, CA 92093, USA
^c Department of Microbiology and Immunology, Stanford University School of Medicine, CSSR3220, 269 Campus Drive, Stanford, CA 94305, USA
^d Division for Experimental Feto-Maternal Medicine, Department of Obstetrics and Fetal Medicine, University Medical Center Hamburg-Eppendorf, Center for Obstetrics and Pediatrics, Martinistrasse 52, 20246 Hamburg, Germany

^∗Corresponding author. Department of Computer Science and Electrical Engineering, Institut Visual and Analytic Computing, University of Rostock, Albert-Einstein-Straße 22, 18059 Rostock, GermanyDepartment of Computer Science and Electrical EngineeringInstitut Visual and Analytic ComputingUniversity of RostockAlbert-Einstein-Straße 22Rostock18059Germany

Résumé

Multiple studies have hinted at a complex connection between maternal stress and preterm birth (PTB). This article describes the potential of computational methods to provide new insights into this relationship. For this, we outline existing approaches for stress assessments and various data modalities available for profiling stress responses, and review studies that sought either to establish a connection between stress and PTB or to predict PTB based on stress-related factors. Finally, we summarize the challenges of computational methods, highlighting potential future research directions within this field.

Le texte complet de cet article est disponible en PDF.

Keywords : Stress, Preterm birth, Machine learning, Artificial intelligence