Modeling Alternating Electric Fields for Brain Tumor Treatment


Alternating electric fields or tumor treating fields (TTFields) therapy, is a type of electromagnetic field therapy to treat cancer, particularly recurrent glioblastoma. TTFields therapy delivers low-intensity electric currents through arrays on the scalp and has demonstrated good results compared to traditional methods. However, the precise distribution of TTFields within the brain, and the extent to which they cover the recurrent glioblastoma, remains poorly understood. Computational simulation from medical image data offers a solution for analyzing the effect of TTFields therapy for the treatment of glioblastoma. Simpleware software was used to generate models for this purpose.


  • MRI data acquired during treatment of patient with recurrent glioblastoma
  • Simpleware software used to generate 3D head model with transducer arrays
  • Mesh exported to COMSOL Multiphysics® to analyze electric fields distribution
  • Results help inform understanding of TTFields therapy


  • This research was supported in part from A Reason to Ride Research Fund.

Thanks to:

MRI Image Acquisition and Tumor Treatment

Researchers aimed to determine the frequency-dependent electric field distribution within a patient’s brain; for this they used co-registered post-gadolinium T1-weighted, T2 and MP RAGE images, together with pre-specified conductivity and relative permittivity values for the cerebral structures. Retrospective neuroimaging analysis of TTFields therapy in a 67-year-old female patient with recurrent glioblastoma in the posterior right frontal brain 6 months after the initial treatment with neurosurgical resection, cranial irradiation with concomitant daily temozolomide, and followed by adjuvant temozolomide. She then received both bevacizumab, infused at a dose of 10 mg/kg every 2 weeks, and TTFields therapy, with which she applied onto her shaved head continuously when possible. The placement of the transducer arrays was based on a personalized computer-generated layout, which is derived from MRI-based morphometric measurements of her head, tumor size, and tumor location. She continued both treatments for 24 months until another site of disease was detected in the lateral border of the right lateral ventricle.

MRI images of the patient’s brain: 6 months after initial treatment, the tumor can be seen in upper slice (A) and lower slice (B); after 24 months a new site of disease was detected at the lateral border of the right lateral ventricle (D), while the primary tumor was stable (C).

Image Processing and Meshing

From the same baseline MRI used for array layout, a 3D model of the head was generated from co-registered MR image datasets using Simpleware ScanIP. Using Simpleware FE, a finite element mesh was generated for each of the segmented head structures including the scalp, skull, dura, cerebrospinal fluid (CSF), supratentorial gray/white matter, ventricles, brainstem, cerebellum, the recurrent glioblastoma, as well as the transducer arrays. The composite finite element model was then exported as a mesh in COMSOL Multiphysics® format. 

Head model with transducer arrays created using Simpleware software

Simulation Results

The meshes were imported into COMSOL Multiphysics® (5.0) to solve for the electric field distribution. The respective conductivity and relative permittivity values used were 0.001 S/m and 1110 for scalp, 0.021 S/m and 204 for skull, 0.502 S/m and 290 for dura, 2.000 S/m and 109 for CSF, 0.141 S/m and 2010 for gray matter, 0.087 S/m and 1290 for white matter, and 1.000 S/m and 10,000 for recurrent glioblastoma. 

Analysis of electric fields in COMSOL Multiphysics®: High field strength is seen at the ventricular horns and there is inhomogeneous electric field distribution on the tumor (A). The new enhancing tumor is situated at the lateral border of the right lateral ventricle in an area with relatively lower electric field strength (B, arrow). 

Analysis revealed valuable information on the procedure, including the location of the highest electric field strength around the ceramic disks for each array, and the potential to reduce scalp irritation. Inhomogeneous electric field distribution was also observed on the tumor, while the MRI revealed a new enhancing tumor subsequent to treatment in a location with relatively low electric field strength. Being able to better understand the distribution of electric currents using models generated from imaging data offers an important method for analyzing the efficacy of TTFields therapies compared to conventional chemotherapies.