Quantitative Analysis of Bladder Wall Thickness for Magnetic Resonance Cystoscopy

• Published in: IEEE transactions on biomedical engineering Vol. 62; no. 10; pp. 2402 - 2409
• Main Authors: Zhang, Xi, Liu, Yang, Yang, Zengyue, Tian, Qiang, Zhang, Guopeng, Xiao, Dan, Cui, Guangbin, Lu, Hongbing
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.10.2015; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: 3D bladder template; Aged; Bladder; bladder cancer; bladder wall segmentation; bladder wall thickness; Cancer; Cystoscopy - methods; Excretory system; Filling; Humans; Image Processing, Computer-Assisted - methods; Laplace equations; Magnetic Resonance Imaging - methods; Male; Pipelines; Shape; shape mapping; T2-weighted MRI; Three-dimensional displays; Urinary Bladder - anatomy & histology; Urinary Bladder - pathology; Urinary Bladder Neoplasms - pathology; bladder wall segmentation; bladder wall thickness (BWT); T2-weighted magnetic resonance image (MRI); 3-D bladder template; Bladder cancer; shape mapping
• ISSN: 0018-9294; 1558-2531
• DOI: 10.1109/TBME.2015.2429612

Objective: To find an effective way for quantitative evaluation on wall thickness variation of human bladder with/without bladder tumor, a novel pipeline of thickness measurement and analysis for magnetic resonance (MR) cystography is proposed. Methods: After the acquisition of volumetric bladder images with a high-resolution T2-weighted 3-D sequence, the inner and outer borders of the bladder wall were segmented simultaneously by a coupled directional level-set method. Then, the bladder wall thickness (BWT) was estimated using the Laplacian method. To reducing the influence of individual variation and urine filling on wall thickness, a thickness normalization using Z-score is performed. Finally, a parametric surface mapping strategy was applied to map thickness distribution onto a unified sphere surface, for quantitative intra- and intersubject comparison between bladders of different shapes. Results: The proposed pipeline was tested with a database composed of MR bladder images acquired from 20 volunteers and 20 patients with bladder cancer. The results indicate that the thickness normalization step using Z-score makes the quantitative comparison of wall thickness quite possible and there is a significant difference on BWT between patients and volunteers. Using the proposed pipeline, we established a thickness template for a normal bladder wall based on dataset of all volunteers. Conclusion: As a first attempt to establish a general pipeline for bladder wall analysis, the presented work provides an effective way to achieve the goal of evaluating the entire bladder wall for detection and diagnosis of abnormality. In addition, it can be easily extended to quantitative analyses of other bladder features, such as, intensity-based or texture features.