MENU

Click here to

×

Are you sure ?

Yes, do it No, cancel

Classification of Urinary Stone Composition with a Photon-Counting-Detector CT System Used in a Simulated Dual-Source Acquisition Mode

G Michalak1*, L Ren2 , A Ferrero3 , S Leng4 , C McCollough5 , (1) ,Rochester, MN, (2) Mayo Clinic, Rochester, MN, (3) Mayo Clinic, Rochester, MN, (4) Mayo Clinic, Rochester, MN, (5) Mayo Clinic, Rochester, MN

Presentations

(Wednesday, 7/17/2019) 7:30 AM - 9:30 AM

Room: 221AB

Purpose: Our purpose was to determine the ability of a photon-counting-detector (PCD) CT system, used in a simulated dual-source mode (sDS PCD-CT), to classify urinary stones and to compare the results to those from a clinical dual-source dual-energy (DSDE) scanner.

Methods: Eighty-six human urinary stones were scanned in 25, 35, and 45cm anthropomorphic water phantoms using a sDS PCD-CT system. Two consecutive scans were performed on each phantom, one using 80 kV and one at 140 kV with a tin filter. Energy thresholds of 25/50 keV and 25/90 keV were applied to the 80 and 140Sn kV beams, respectively. Stones were classified based on the ratio of the CT numbers in the lowest and highest energy bins. Receiver-operating-characteristic curves were generated for each phantom size and classification task, and the area under the curve (AUC) was used as the figure of merit. Classification was performed for uric acid (UA) vs. non-uric acid (non-UA) stones, as well as the non-UA subtypes, calcium oxalate (COX) vs. apatite (APA).

Results: In the 25cm phantom, the AUC’s for classification of UA vs. non-UA and COX vs. APA were 1 and 0.846, respectively. When compared with a DSDE scanner evaluated in a previous study (Siemens FLASH; 80/140Sn for 35cm and 100/140Sn for 45cm), sDS PCD-CT classification performed equally well in the 35cm phantom (AUC = 1 for both) and superiorly in the 45 cm phantom (AUC = 1 vs 0.923) for UA vs. non-UA. sDS PCD-CT classification was superior for COX vs. APA in the 35cm phantom (AUC = 0.849 vs. 0.684) and the 45 cm phantom (AUC = 0.689 vs. 0.588).

Conclusion: Considering both classification tasks and phantom sizes, sDS PCD-CT provides superior classification performance compared to DSDE CT.

Funding Support, Disclosures, and Conflict of Interest: This research was funded by Siemens Healthcare.

Keywords

Not Applicable / None Entered.

Taxonomy

Not Applicable / None Entered.

Contact Email