Comparison of bladder volume measurement accuracy between two-dimensional ultrasound with three-dimensional reconstruction and conventional two-dimensional ultrasound
10.3760/cma.j.cn113030-20250227-00072
- VernacularTitle:二维超声三维重建法与常规二维超声测量膀胱容量准确性比较
- Author:
Kaixuan ZHANG
1
;
Ying CAO
1
;
Lijing ZUO
1
;
Zhen WANG
1
;
Wensheng NIE
1
;
Yongli SONG
1
;
Xing LIU
1
;
Mingjian SUN
1
;
Yuan TANG
1
;
Yueping LIU
1
Author Information
1. 国家癌症中心/国家肿瘤临床医学研究中心/中国医学科学院北京协和医学院肿瘤医院放疗科，北京 100021
- Publication Type:Journal Article
- Keywords:
Tomography;
Bladder ultrasonography;
X-ray computed;
Spiral computed tomography;
Radiotherapy;
Pelvic neoplasms;
Three-dimensional reconstruction
- From:
Chinese Journal of Radiation Oncology
2025;34(12):1238-1244
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To compare the accuracy of two-dimensional (2D) ultrasound with three-dimensional (3D) reconstruction and conventional 2D ultrasound in measuring bladder volume in pelvic tumor patients, using computed tomography (CT) as the reference.Methods:A set of bladder phantoms were constructed to compare CT and ultrasound measurements with actual injected volumes. Clinical data of 104 pelvic tumor patients who received radiotherapy at the Cancer Hospital, Chinese Academy of Medical Sciences between August and December 2023 were retrospectively analyzed. Portable transabdominal ultrasound was used to obtain the largest bladder cross-section, and the maximum diameters in the left-right (LR), anterior-posterior (AP), and superior-inferior (SI) directions (D LR, D AP, D SI) were measured. The 2D ultrasound volume was calculated as V=0.523 × D LR × D AP × D SI. Full-bladder transverse videos were recorded and processed in Matlab R2016a through frame extraction(60 images), followed by contrast enhancement, edge detection segmentation, cubic spline interpolation, and image smoothing to achieve 3D reconstruction. Paired t-tests, intraclass correlation coefficients (ICC), and Bland-Altman analyses were performed to assess systematic bias and consistency between ultrasound methods and CT. Multivariate linear regression was applied to evaluate the effects of slice thickness, posture, age, and other factors on CT measurements. Results:In the phantom study, deviations of 2D ultrasound and CT from actual injected volumes were (0.73±3.05) ml ( t=-0.48, P=0.667) and (1.52±11.27) ml ( t=0.17, P=0.875), with ICC values>0.999. In the clinical study, mean bladder volumes measured by 3D-reconstructed ultrasound, conventional 2D ultrasound, and CT were (373.5±153.31), (314.89±135.28), (382.82±157.57) ml, respectively. The 3D-reconstructed method showed excellent agreement with CT (ICC=0.98; Bland-Altman mean bias=-9.32 ml, P=0.096), while 2D ultrasound also showed good consistency (ICC=0.91), but significantly underestimated bladder volume (mean bias=-67.93 ml, P<0.001). Subgroup analysis revealed that 2D ultrasound had the best agreement with CT in the medium-volume group (200-500 ml, ICC=0.902), whereas agreement decreased in the small-volume (<200 ml, ICC=0.884) and large-volume (>500 ml, ICC=0.840) groups (all P<0.001). The 3D-reconstructed ultrasound maintained excellent consistency with CT across all subgroups (all ICC>0.95), and the measured bladder volume was not statistically significant. Multivariate regression showed that slice thickness, posture, age, sex, and surgical status had no significant effects on CT measurements. Conclusions:Ultrasound with 3D reconstruction enables accurate bladder volume monitoring through true 3D contour reconstruction, while conventional 2D ultrasound systematically underestimates bladder volume and requires correction.
