Validation of digital radiology measurement tools for quantitative spinal imaging
Background/Objective: Digital imaging tools for linear and angular measurements facilitate measurement requirements compared to conventional techniques. However, the accuracy of these measurement, especially in spinal trauma, has nol been assessed. The objective of this study was to validate current onscreen measurement tools available on picture archiving and communication systems (PACS) tor distance and angular measurements in cervical spine plain digital radiographs and computed tomographic (CT) images. Method: A cervical spine sawbone was prepared with radio-opaque markers at known distances and angles. This model was lhen imaged with digital plain radiographs (spine model at a fixed [24 cm] and variable distance from the X-ray receiver) and CT (1-mm slice thickness) at 4 centers. Using 4 different commercially available digital imaging viewing programs, linear and angular measurements were performed by a musculoskeletal radiologist and 5 spinal surgeons. Results: At only 24 cm (distance from the anatomic center of the spine to lateral shoulder of an average 70-kg male) from the receiver, linear measurements were 5-8 mm (∼40%) greater then the actual distance (p < .0001). This error varied with the distance between the receiver and spine model (i.e., degree of magnification) and was correctable by internal calibration of the measurement tool with an object of known dimension. Linear measurements from CT were accurate to within 1 mm of the actual measurement. Angular measurements were not affected by magnification. Conclusion: The results of this study show that linear measurements obtained on a PACS system are valid for CT but grossly invalid for digital plain X-rays without the use of an internal calibration reference. © 2006 Thomas Land Publishers, Inc.
Rampersaud, YR; Fehlings, MG; Harrop, JS; Kuklo, TR; Massicotte, EM; Salonen, D; Shaffrey, CI; Vaccaro, AR
Volume / Issue
Start / End Page
International Standard Serial Number (ISSN)
Digital Object Identifier (DOI)