Finite element based bladder modeling for image-guided radiotherapy of bladder cancer.

Research output: Contribution to journalArticle

  • External authors:
  • Xiangfei Chai
  • Marcel van Herk
  • Jeroen B van de Kamer
  • Maarten C C M Hulshof
  • Peter Remeijer
  • Heidi T Lotz
  • Arjan Bel

Abstract

PURPOSE: A biomechanical model was constructed to give insight into pelvic organ motion as a result of bladder filling changes. METHODS: The authors used finite element (FE) modeling to simulate bladder wall deformation caused by urine inflow. For ten volunteers, a series of MRI scans of the pelvic area was recorded at regular intervals of 10 min over 1 h. For the series of scans, the bladder volume gradually increased while the rectal volume was constant. The MR image with the bladder volume closest to 250 ml was selected as the reference in each volunteer. All pelvic structures were defined from the reference image including bladder wall, small bowel, prostate (male), uterus (female), rectum, pelvic bone, and the rest of the body. These structures were translated to FE meshes. Using appropriate material properties for all organs, deformations of these organs as a response to changing bladder pressure were computed. RESULTS: The computation results showed realistic anisotropic deformation of the bladder wall: The bladder became more elongated in the cranial and anterior directions with increasing bladder volume. After fitting the volume of the computed bladder to the actual bladder volume on the test images, the computed bladder shape agreed well with the real bladder shape (overlap from 0.79 to 0.93). The average mean bladder wall prediction errors of all the volunteers were 0.31 cm average and 0.29 cm SD. CONCLUSIONS: In conclusion, a FE based mechanical bladder model shows promise for the prediction of the short-term bladder shape change using only one pelvic scan and volume change of the bladder as input. The accuracy levels achieved with this method are likely mostly limited by inaccuracies in material properties and sliding tissue between organs, which has not been modeled. This model can potentially be used to improve image-guided radiotherapy for bladder cancer patients, i.e., by prediction short-term bladder deformation.

Bibliographical metadata

Original languageEnglish
JournalMedical Physics
Volume38
Issue number1
Publication statusPublished - Jan 2011