Login to Access Video or Poster Abstract: MP18-17
Sources of Funding: None

Introduction

Stone treatment decisions are guided by stone size and stone density. We have previously shown that automated stone volume measurements are more precise than manual size measurements. We tested a novel dedicated renal stone software program that provides a comprehensive radiographic stone profile with a single click._x000D_

Methods

Urinary stones identified on CT scans were measured by a single reader to obtain the longest length on axial images and coronal images. A region of interest (ROI) was drawn inside of the stone to obtain minimum, maximum, average and standard deviation densities. The same stones were then assessed with an automated comprehensive radiographic stone profile software. This software produces the minimum and maximum linear diameter; minimum, maximum, and average stone density as well as stone volume. Maximum linear diameter, maximum density and average densities were compared between the manual measurements and the automated software data to obtain the percent difference between the measurements. The longer of either the axial or coronal manual measurements was used to compare with the maximum linear diameter from the stone profile software.

Results

17 patients were identified who had a total of 42 CT scans with 85 unique stones. Patients had an average of 2.5 scans with an average of 5 stones. Stone sizes ranged from 1.9 mm to 21 mm in length with a mean of 8 mm. The average density measurement was 451 (21-1492). Volume obtained from the stone profile software averaged 182 mm3 (2.8-2668 mm3). The maximum diameter between the manual and software differed by an average of 19.1% (0-54.8%). Maximum density differed by an average of 11% (0-66.6%) and average density differed by an average of 24% (0-78%).

Conclusions

Automated comprehensive radiographic stone profiling can be accomplished with a single click and closely approximates manual measurements which can be laborious, requiring the reader to measure length on 2 different CT scan reconstructions and carefully place a region of interest around the stone. Automated stone profiling was able to assess stones as small as 1.9mm as well as stones with complex geometry. Stone volume was obtained in true maximum length, and density information thus facilitating rapid and accurate “one click measurements� for clinical decision making.

Funding

None