An image processing approach to compensate for the bladder wall motion and deformation in MR cystography

Bladder cancer becomes the 5th leading cancer incidence. Management of bladder tumor is difficult because of its high recurrence rate (as high as 80%) after resection. It causes tremendous suffering and stress to the patient and the family. A non-invasive means of evaluating the entire bladder wall, especially the wall thickness as a biomarker, is clinically desired. MRI can provide such a means because of the intrinsic contrast between the urine and the wall by T
₁
and T
₂
relaxations. However, an acquisition without signal average (short-time scan) suffers from low signal-to-noise ratio (SNR), while with signal average (long-time scan) introduces motion artifacts because of the bladder wall deformation due to urine inflow, intestines' peristalsis and lung breathing, etc. To enhance the SNR while mitigating the motion artifacts, we proposed an image-processing approach of acquiring repeated short-time scans without signal average and registering the repeated scans prior to average. The effectiveness of the approach depends on the noise level in the short-time scans and the robustness of the registration method. This pre-requisition leads us to investigate adequate pulse sequences for fully 3D data acquisition and registration method. After investigating several potential pulse sequences, we found the THRIVE pulse sequence with TR=6.23ms, TE=3.87ms, sampling size of 208×100, and flip angle of 10 degrees produced adequate tradeoff of noise and contrast in images of 224×224 array with voxel size of 0.9375×0.9375×1.0 mm
³
. Then a group-wise B-spline based 3D deformable algorithm was developed to register the short-time scans for compensating the bladder wall motion. Results from volunteer and patient studies demonstrated the effectiveness of the presented simple, innovative approach in reducing bladder motion, enhancing the SNR and preserving edge details for imaging the bladder wall and, therefore, facilitating the measure of the wall thickness and detecting bladder wall abnormality.