Postoperative low-dose follow-up computed tomography for patients with stomach cancer: optimal blending ratio of adaptive statistical iterative reconstruction for image quality and diagnostic performance.

OBJECTIVE

To compare the diagnostic performance of adaptive statistical iterative reconstruction applied low-dose computed tomography (CT) (LDCT) with that of the standard-dose CT (SDCT) for local recurrence in patients with stomach cancer.

METHODS

Seventy-nine consecutive patients who had undergone surgical resection for stomach cancer were enrolled. To monitor recurrence, SDCT (120 kilovolt peak [kVp], 200mAs) had been performed. The LDCT (120 kVp, 100 mA s) was taken, and images were reconstructed with 4 levels of adaptive statistical iterative reconstruction (ASIR) blending (0%, 30%, 50%, and 70%). Two blinded radiologists recorded the diagnostic confidence scores for local recurrence in each data set using a 5-point scale. Endoscopic biopsy results served as the reference standard. Receiver operating characteristic (ROC) curve analysis was used to calculate the diagnostic performance.

RESULTS

The diagnostic performance of LDCT with variable ASIR blending ratios was comparable to that of SDCT (area under ROC curve, 0.727-0.734, 0.687, respectively, P > 0.05).

CONCLUSIONS

The diagnostic performance of ASIR applied LDCT is comparable to that of SDCT.