Analysis of breathing via optoelectronic systems: comparison of four methods for computing breathing volumes and thoraco-abdominal motion pattern

摘要

Breathing parameters can be measured by motion capture systems by placing photo-reflective markers on the chest wall. A computational model is mandatory to compute the breathing volume and to calculate temporal and kinematical features by the gathered markers trajectories. Despite different methods based on different geometrical approaches can be adopted to compute volumes, no information about their differences in the respiratory evaluation are available. This study investigated the performances of four methods (conventional, prism-based, convex hull with boundary condition, based on Delaunay triangulation) using an optoelectronic motion capture system, on twelve healthy participants during 30 s of breathing. Temporal trends of volume traces, tidal volume values, and breathing durations were compared between methods and spirometry (used as reference instrument). Additionally, thoraco-abdominal motion patterns were compared between methods by analysing the compartmental contributions and their variability. Results shows comparable similarities between the volume traces obtained using spirometry, prism-based and conventional methods. Prism-based and convex hull with boundary condition methods show lower bias in tidal volumes estimation up to 0.06 L, compared to the conventional and Delaunay triangulation methods. Prism-based method shows maximum differences of 30mL in the comparison of compartmental contributions to the total volume, by resulting in a maximum deviation of 1.6% in the percentage contribution analysis. In conclusion, our finding demonstrated the accuracy of the non-invasive MoCap-based breathing analysis with the prism-based method tested. Data provided in this study will lead researchers and clinicians in the computational method choice for temporal and volumetric breathing analysis.