Quantitative videobronchoscopy: a new technique to assess airway caliber.

Quantitative assessment of airway caliber is generally confined to indirect physiologic methods or to radiographic techniques. Fiberoptic bronchoscopy provides a direct view of airways, permitting quantification of airway caliber by image analysis. We investigated the characteristics of a bronchoscopic imaging system, determined its limitations in quantification and the corrections necessary for accurate assessment of image dimension, validated the methodology with airway models, and applied the technique to airways in vivo. The system comprised a bronchoscope, videocamera, videocassette recorder (VCR), computer with a frame grabber, and image-analysis program. Image quantification was affected by two sources of distortion: (1) Distance distortion: a loss of image resolution with increasing distance between the object and bronchoscope, requiring determination of the operational distance range. (2) Radial distortion: a progressive reduction in image size from the center to the periphery of the bronchoscopic field of view (FOV), requiring correction of airway dimension according to airway size and location in FOV. Validation of the methodology with different sized airway models indicated an underestimation of measured diameters, which normalized with distortion correction. We provide an example of quantitative videobronchoscopy with measurements of in vivo airway narrowing due to vagal stimulation in the anesthetized dog. Measurements of airway narrowing made with videobronchoscopy were also compared with those made with high-resolution computer-assisted tomography (HRCT) which suggested that the two technologies provide unique but complementary perspectives on airway dimensions. We conclude that videobronchoscopy and image analysis provide a novel and accurate method for the quantification of airway caliber.