Accuracy of clavicle bone measurement in detection of patient rotation during chest radiography; a cross-sectional study

Patient rotation during chest x-ray is one of the important causes of decreased image quality and accuracy. Therefore, this study was designed and performed to calculate the accuracy of measuring the length of clavicle bones in detection of patient rotation during chest radiography.

In this diagnostic study, chest radiographs of 435 patients referring to Loghman Hakim Hospital were investigated for rotation via three

standard method (by a radiologist), measurement of clavicle bones (by emergency medicine resident), and visual method (by emergency medicine specialist). The information was recorded in a checklist and sensitivity, specificity, and positive and negative predictive values of the last two methods were calculated in comparison with the standard method for detecting patient rotation.

In comparison with the standard method, the method of measuring the length of the clavicle had a sensitivity of 88.9%, specificity of 97.6%, a positive predictive value of 99.4%, a negative predictive value of 67.8%, and an accuracy of 90.6%. The visual method had a sensitivity of 66.9%, a specificity of 71.4%, a positive predictive value 90.7%, a negative predictive value 34.1%, and an accuracy of 67.8% compared to the standard method in determining the presence or absence of chest rotation.

Overall, the findings of our study indicated that the accuracy of clavicle length measurement method is not much different from that of the standard method in detection of patient rotation and can be used as its replacement in this regard.