Volume 24, Issue 101, January - February, 2020

Diagnosis of splenomegaly with one-dimensional diameters in CT scan images in North-East of Iran and constructing a best-fit model for estimation of splenic volume

Masoud Pezeshki Rad¹, Maryam Salehi², Mohaddeseh Mohammadi^3♦, Fatemeh Soati⁴

¹Department of Radiology, Faculty of Medicine, Mashhad University of Medical Sciences, Mashhad, Iran Email: pezeshkiradm@mums.ac.ir
²Department of Community Medicine, Faculty of Medicine, Mashhad University of Medical Sciences, Mashhad, Iran Email: salehim@mums.ac.ir
³Department of Radiology, Faculty of Medicine, Golestan University of Medical Sciences, Golestan, Iran Email: mohaddese_mohammadi@yahoo.com
⁴Department of Radiology, Faculty of Medicine, Tehran University of Medical Sciences, Tehran, Iran Email: arezoosoati@yahoo.com

^♦Corresponding author
Department of Radiology, Faculty of Medicine, Golestan University of Medical Sciences, Golestan, Iran Email: mohaddese_mohammadi@yahoo.com

ABSTRACT

Introduction: The purpose of present study was to evaluate the accuracy rate of diagnosis of splenomegaly with one-dimensional diameters in computed tomography (CT) scan images and constructing a best-fit model for estimation of splenic volume from the standardized one-dimensional diameters of the spleen to estimates splenic volume with the prolate ellipsoid and a best-fit model. Materials and Methods: This cross-sectional study was performed retrospectively on 200 patients’ abdominal CT scan images (100 with splenomegaly and 100 without splenomegaly). Six one-dimensional diameters were registeredfor each abdominal CT scan; splenic length (L), width (W) and four sizes of thickness (T1–T4). The volume of spleen was calculated by the following formula (volume= W× T3× π/6) and also by MITK software. Also the formula for splenic volume was obtained by using linear regression. Sensitivity and specificity values were obtained for the one-dimensional index. Using ROC curve, the highest sensitivity and specificity for splenic length, width and thickness was determined to achieve a cutoff point for definition of splenomegaly unidimensionally. Results: In our study the length of 11.33 cm was the most sensitive and specific uni-dimensional parameter to define splenomegaly (sensitivity = 85% and specificity = 89%).The best regression model for predicting splenic volume was calculated Volume=0.4× (L×W×T3) +5 with R2=%90. Conclusion: We constructed this formula (0/4× (W×T3×L) +5) with less bias for estimation of splenic volume.

Keywords: Computerized Tomography scan; Splenic Volume; Volumetry

Medical Science, 2020, 24(101), 66-74

PDF