July 29, 2024Scientific ReportsOpen Access

Transitioning sleeping position detection in late pregnancy using computer vision from controlled to real-world settings: an observational study

View Full Paper

Ask AI

Bookmark

Authors

Discussion

Observational study demonstrates automated sleeping position detection in late pregnancy, suggesting feasibility for real-world maternal monitoring.

Sleeping position detection achieved an 82-to-89% average probability using deep convolutional neural networks, with a 15-to-19% chance of missed detection.
Observational study using five-fold cross validation evaluated 12,930 video images from 39 pregnant participants across a four-night home sleep apnea study.
Supports real-world computer vision monitoring to assess late stillbirth risks, helping clarify causal links with small-for-gestational-age infant delivery.

Kember et al. (2024) studied this question.

View Full Paper

Ask AI

Bookmark

Observational study demonstrates automated sleeping position detection in late pregnancy, suggesting feasibility for real-world maternal monitoring.