Non-Contact Physiological Monitoring in Pediatric Intensive Care Units via Adaptive Masking and Self-Supervised Learning

2026-02-17

AI Summary

This study addresses the limitations of current non-contact remote photoplethysmography (rPPG) methods in pediatric intensive care, where motion artifacts, occlusions, and illumination variations compromise clinical reliability, while contact-based monitoring risks skin injury and discomfort. To overcome these challenges, the authors propose a novel self-supervised rPPG estimation framework that integrates a VisionMamba architecture with an adaptive masking mechanism. A lightweight Mamba controller dynamically assigns spatiotemporal importance scores to guide probabilistic video patch sampling, enabling the model to focus on pulse-rich regions without explicit region-of-interest extraction. The approach employs a three-stage curriculum learning strategy combined with teacher–student knowledge distillation. Evaluated on videos from 500 pediatric patients, the method achieves a mean absolute error (MAE) of 3.2 bpm—42% lower than standard masked autoencoders and 31% better than PhysFormer—while maintaining robustness under occlusion and noise.

Technology Category

• Computer Vision: Multi-modal Vision
• Intelligent Robots: State Estimation
• Machine Learning: Hardware-aware ML

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Continuous monitoring of vital signs in Pediatric Intensive Care Units (PICUs) is essential for early detection of clinical deterioration and effective clinical decision-making. However, contact-based sensors such as pulse oximeters may cause skin irritation, increase infection risk, and lead to patient discomfort. Remote photoplethysmography (rPPG) offers a contactless alternative to monitor heart rate using facial video, but remains underutilized in PICUs due to motion artifacts, occlusions, variable lighting, and domain shifts between laboratory and clinical data. We introduce a self-supervised pretraining framework for rPPG estimation in the PICU setting, based on a progressive curriculum strategy. The approach leverages the VisionMamba architecture and integrates an adaptive masking mechanism, where a lightweight Mamba-based controller assigns spatiotemporal importance scores to guide probabilistic patch sampling. This strategy dynamically increases reconstruction difficulty while preserving physiological relevance. To address the lack of labeled clinical data, we adopt a teacher-student distillation setup. A supervised expert model, trained on public datasets, provides latent physiological guidance to the student. The curriculum progresses through three stages: clean public videos, synthetic occlusion scenarios, and unlabeled videos from 500 pediatric patients. Our framework achieves a 42% reduction in mean absolute error relative to standard masked autoencoders and outperforms PhysFormer by 31%, reaching a final MAE of 3.2 bpm. Without explicit region-of-interest extraction, the model consistently attends to pulse-rich areas and demonstrates robustness under clinical occlusions and noise.

Problem

• non-contact physiological monitoring
• pediatric intensive care
• remote photoplethysmography
• motion artifacts
• domain shift

Innovation

• self-supervised learning
• adaptive masking
• VisionMamba
• remote photoplethysmography (rPPG)
• curriculum learning