Premium
A combined registration and finite element analysis method for fast estimation of intraoperative brain shift; phantom and animal model study
Author(s) -
Mohammadi Amrollah,
Ahmadian Alireza,
Rabbani Shahram,
Fattahi Ehsan,
Shirani Shapour
Publication year - 2017
Publication title -
the international journal of medical robotics and computer assisted surgery
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.556
H-Index - 53
eISSN - 1478-596X
pISSN - 1478-5951
DOI - 10.1002/rcs.1792
Subject(s) - finite element method , image registration , imaging phantom , computer science , estimator , computation , process (computing) , deformation (meteorology) , algorithm , artificial intelligence , mathematics , image (mathematics) , physics , nuclear medicine , structural engineering , medicine , statistics , engineering , meteorology , operating system
Background Finite element models for estimation of intraoperative brain shift suffer from huge computational cost. In these models, image registration and finite element analysis are two time‐consuming processes. Methods The proposed method is an improved version of our previously developed Finite Element Drift (FED) registration algorithm. In this work the registration process is combined with the finite element analysis. In the Combined FED (CFED), the deformation of whole brain mesh is iteratively calculated by geometrical extension of a local load vector which is computed by FED. Results While the processing time of the FED‐based method including registration and finite element analysis was about 70 s, the computation time of the CFED was about 3.2 s. The computational cost of CFED is almost 50% less than similar state of the art brain shift estimators based on finite element models. Conclusions The proposed combination of registration and structural analysis can make the calculation of brain deformation much faster.