Quantitative Analysis by 3D Graphics of Thoraco-Abdominal Surface Shape and Breathing Motion

Abstract

Chest wall motion can provide information on respiratory muscles' action and on critical vital signs, like respiration and cardiac activity. The chest wall is a structure with three compartments that are independent to each other and can move paradoxically according to the pathophysiology of the disease. Opto-electronic plethysmography (OEP) allows for non-invasively 3D tracking of body movements. We aimed to extend the characteristics of OEP analysis to local analyses of thoraco-abdominal surface geometry and kinematics during respiration. Starting from the OEP output file, the 3D markers’ coordinates were combined with a triangulation matrix. A smoothing procedure (an automatic and iterative interpolation process to increase the number of vertices from 93 to 548) was applied to allow for precise local analysis of the thoraco-abdominal surface. A series of measurements can be performed to characterize the geometry of the trunk and its three compartments, in terms of volumes, height, diameters, perimeters, and area. Some shape factors, such as surface-to-volume ratio or height-to-perimeter ratio, can be also computed. It was also possible to build the vector field associated with the breathing motion of all the vertices, in terms of magnitude and motion direction. The vector field data were analyzed and displayed through two graphic tools: a 3D heatmap, in which the magnitude of motion was associated to different colors, and a 3D arrow plot, that allowed us to visualize both the magnitude and the direction of motion with color-coded arrows. The methods were applied to 10 healthy subjects (5 females) and also applied to two cases: a pregnant woman at each trimester of gestation and a patient before and after a demolition thoracic surgery. The results proved to be coherent with the physiology of healthy subjects and the physiopathology of the cases. We developed a new non-invasive method for respiratory analysis that allowed for the creation of realistic 3D models of the local and global trunk surface during respiration. The proposed representation constituted a very intuitive method to visualize and compare thoraco-abdominal surface movements within and between subjects, therefore enforcing the potential clinical translational value of the method.