Facial 3D data acquisition in critically ill children for production of personalized non-invasive ventilation masks

Abstract

Background: Non-invasive ventilation is commonly used to support critically ill children with acute respiratory failure in the pediatric intensive care unit. However, non-invasive ventilation treatment is often hindered by poorly fitting masks due to limited commercially available options. Personalized non-invasive ventilation masks are a promising solution, yet research on the feasibility of their production in real-world clinical settings, particularly regarding facial data acquisition, remains limited. This study aims to assess the feasibility of using a handheld 3D scanner for facial data acquisition in critically ill children admitted to the pediatric intensive care unit. Methods: In this single-center pediatric intensive care unit feasibility study, facial 3D data was obtained from children (age 0–18 years) receiving non-invasive respiratory support for acute respiratory failure, using a handheld 3D scanner. Feasibility outcomes included the scan process and quality factors. Scan quality was evaluated based on scan errors and removed movement frames. Facial 3D data acquisition was defined as feasible if > 80% of patients had a complete scan whereof > 90% frames had a scan error < 0.5. Results: We included 33 patients with a median (IQR) age of 2.0 (1.0–16.0) months. Full facial 3D data could be acquired within a short scanning period of 30 s, which did not induce patient clinical deterioration, with a success rate of 31 (94%) usable scans with good quality (98% good frames). Conclusion: Our results show that facial data acquisition using a handheld 3D scanner is feasible in critically ill children receiving non-invasive respiratory support in the pediatric intensive care unit. These findings are essential for developing and implementing a workflow process for personalized non-invasive ventilation masks for children with acute respiratory failure.