Click here to


Are you sure ?

Yes, do it No, cancel

Verification of Dose Calibrator Consistency for Tc-99m XSPECT System Calibration in a Large Clinical Nuclear Medicine Department

M Jacobsen*, W Erwin , UT MD Anderson Cancer Center, Houston, TX


(Monday, 7/15/2019) 9:30 AM - 10:00 AM

Room: Exhibit Hall | Forum 8

Purpose: xSPECT (Siemens Healthineers) generates images with units of Bq/ml or standardized uptake value. xSPECT requires dose calibrators to be individually cross-calibrated with each scanner and a NIST-traceable Co-57 source, which is challenging for large enterprises that use multiple dose calibrators. The consistency of measurements across various dose calibrators was evaluated, to determine whether a single xSPECT cross-calibration could be applied to multiple dose calibrators.

Methods: Two xSPECT Co-57 calibration sources (Eckert & Ziegler) and ten Tc-99m bone doses were assayed in five dose calibrators (Capintec CRC-15R) using factory settings for Co-57 and Tc-99m. Each xSPECT source was assayed twice per dose calibrator, while the ten Tc-99m sources were each assayed once per dose calibrator. Percent error for each dose calibrator was calculated using the Co-57 sources. Coefficients of variation (CoVs) were calculated for each Tc-99m and Co-57 source to assess the variability between dose calibrators during routine clinical use.

Results: The average difference between the measured and time-corrected known Co-57 activity on a given dose calibrator ranged from 0.19% to 0.73%, with an average of 0.40%, which is on the order of the 3% uncertainty (k=3) in the source activity. The CoVs in each Co-57 source were 0.42% and 0.32% across the five dose calibrators. The Tc-99m sources’ CoVs across dose calibrators ranged from 0.57% to 1.47%. The maximum absolute difference between dose calibrator output for a single source (reported as a percentage of the average reading for a given source), ranged from 1.3% to 3.3%.

Conclusion: Dose calibrators of the same model provided consistent activity measurements with errors relative to the known activity below 1% for NIST-traceable Co-57 sources, and CoVs across dose calibrators well below 5%. Since 10% is considered an acceptable upper limit on dose calibrator accuracy, cross-calibration with one dose calibrator may suffice.


Nuclear Medicine, SPECT


IM- Nuclear Medicine General: radionuclide calibration

Contact Email