Clinical Evaluation of an AI-Based Prototype for Contactless Respiratory Monitoring in Children

Ludwig Maximilian Seebauer1, Marcel Geis2, Niklas Alexander Köhler2

  • 1Department of Pediatric Pneumology and Allergy, University Children's Hospital Regensburg (KUNO) at the Hospital St. Hedwig of the Order of St. John, University of Regensburg, D-93049 Regensburg, Germany.

PubMed

Insights

A new contactless video prototype accurately monitors pediatric respiratory rate and abnormal breathing patterns during sleep. This non-invasive method reduces patient burden compared to traditional polysomnography (PSG).

Area of Science:

  • Biomedical Engineering
  • Pediatric Sleep Medicine
  • Respiratory Physiology

Background:

  • Pediatric respiratory disorders require sleep evaluation.
  • Traditional polysomnography (PSG) is resource-intensive and uncomfortable for children.
  • A prior study developed and validated a contactless video prototype for respiratory monitoring.

Purpose of the Study:

  • Clinically validate a contactless monitoring prototype in pediatric patients.
  • Focus on accurate detection of respiratory rate.
  • Identify abnormal pediatric breathing patterns during sleep.

Main Methods:

  • Recruited 27 pediatric patients (6 months-12 years).
  • Used a time-of-flight camera and 3D imaging with AI for contactless thoracoabdominal movement monitoring.
  • Compared prototype-derived respiratory rates with simultaneous PSG data across various conditions and sleep stages.

Main Results:

  • Acquired 296 hours of respiratory data.
  • Analyzed 60-second segments during N3 and REM sleep.
  • Demonstrated feasibility of contactless respiratory monitoring in children.

Conclusions:

  • The contactless prototype enables reliable, non-invasive respiratory monitoring in pediatric patients.
  • Accurately detects respiratory rate and abnormal breathing patterns under clinical conditions.
  • Reduces patient burden and shows potential for clinical adoption.
Abstract