AI-Based Thermal Video Analysis in Privacy-Preserving Healthcare: A Case Study on Detecting Time of Birth

2025-02-05

AI Summary

In clinical practice, neonatal time-of-birth (ToB) is manually recorded with only minute-level precision, insufficient for timely resuscitation—approximately 10% of newborns require assisted ventilation and 5% need positive-pressure ventilation. Method: We propose the first AI system for automatic, privacy-preserving ToB detection from thermal video, eliminating reliance on identity-specific features (e.g., facial or biometric recognition). Our approach integrates temporal thermal-video modeling, a lightweight motion-triggered detection network, and multi-frame confidence-weighted localization. Contribution/Results: Evaluated in real-world delivery settings, the system successfully detected ToB in 96% of cases, achieving a median absolute error of 1 second, 91.4% precision, and 97.4% recall. This work overcomes the limitations of manual recording and establishes the first application of thermal imaging for high-accuracy, non-invasive ToB estimation—providing a reliable temporal anchor for neonatal resuscitation.

Technology Category

• Computer Vision: Biometrics, Face, Gesture & Pose
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Intelligent Robots: State Estimation

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Security and Privacy: Large-scale security measurements

Abstract

Approximately 10% of newborns need some assistance to start breathing and 5% proper ventilation. It is crucial that interventions are initiated as soon as possible after birth. Accurate documentation of Time of Birth (ToB) is thereby essential for documenting and improving newborn resuscitation performance. However, current clinical practices rely on manual recording of ToB, typically with minute precision. In this study, we present an AI-driven, video-based system for automated ToB detection using thermal imaging, designed to preserve the privacy of healthcare providers and mothers by avoiding the use of identifiable visual data. Our approach achieves 91.4% precision and 97.4% recall in detecting ToB within thermal video clips during performance evaluation. Additionally, our system successfully identifies ToB in 96% of test cases with an absolute median deviation of 1 second compared to manual annotations. This method offers a reliable solution for improving ToB documentation and enhancing newborn resuscitation outcomes.

Problem

• Automate Time of Birth detection
• Enhance newborn resuscitation accuracy
• Preserve privacy in healthcare imaging

Innovation

• AI-driven thermal video analysis
• Privacy-preserving healthcare technology
• Automated Time of Birth detection