Objective    To optimize and standardize the technical parameter of multi-slice spiral CT pulmonary perfusion imaging

then to evaluate the effectiveness and stability of this scan protocol.     Methods    First

to determine the scan protocol according to the results of preliminary experiment and literature review

then based on the previous study

we studied in a larger sample size. We scanned 65 cases of lung single nodule according to the scan protocol of multi-slice spiral CT pulmonary perfusion imaging

then analyzed the main factors concerning the imaging quality

evaluated the effectiveness of scan protocol by determining the ROC curve

diagnostic efficacy and odds ratio

as well as the stability of scan protocol by consistency analysis.     Results    Forty-seven cases showed optimal images

resulting in correct diagnosis proved pathologically. Nine cases were made indefinite diagnosis. Nine cases were not up to the image quality for definite diagnosis mainly because of the patients’ incompliant respiration. The total sensitivity

specificity

accuracy

likelihood ratio and the area under ROC curve for scan protocol were 76.9%

83.3%

82%

4.620

0.280

0.840 respectively and Kappa value was 0.894. The diagnostic efficacy of multi-slice spiral CT pulmonary perfusion was significantly higher than that by CT plain and enhanced scan.     Conclusion    The optimized and standardized scan protocol of multi-slice spiral CT pulmonary perfusion imaging is effective in pathologic diagnosis and of well stability

worth being popularized.