A new method for predicting uric acid composition in urinary stones using routine single-energy CT.

Area of Science:

• Radiology
• Urology
• Medical Imaging

Background:

• Urinary stones, particularly uric acid (UA) stones, require accurate identification for effective medical treatment.
• Unenhanced computed tomography (CT) is a common diagnostic tool for urinary stones.
• Distinguishing UA stones from other types using CT is crucial for treatment planning.

Purpose of the Study:

• To analyze quantitative CT parameters from single-energy, thin-slice CT images of urinary stones.
• To correlate these parameters with the chemical composition of the stones.
• To develop a method for differentiating pure UA stones from non-UA/mixed stones.

Main Methods:

• Retrospective analysis of 126 urinary stones (22 UA, 104 non-UA/Mix) from 117 patients using unenhanced, thin-slice, single-energy CT.
• Quantitative analysis of CT images and Laplacian filtered images using operator-independent methods.
• Development of a classification method based on peak attenuation and peak point Laplacian values.

Main Results:

• Significant differences in quantitative image characteristics (mean attenuation, peak attenuation, peak point Laplacian) were observed between UA and non-UA/Mix stones (p < 0.001).
• The developed method achieved 95% sensitivity and 99% specificity in classifying pure UA stones.
• Results demonstrated comparability to dual-energy CT methods.

Conclusions:

• Quantitative image analysis of thin-slice, single-energy CT data shows promise for predicting pure UA content in urinary stones.
• This non-invasive method can aid in distinguishing UA stones, facilitating appropriate medical management.
• The findings support the use of routine CT for stone composition prediction.