Tumor Treating Fields (TTFields) are an antimitotic treatment against brain and other tumors. They are applied regionally and non-invasively by inducing intermediate frequency (100-300 kHz) alternating electric field of intensities between 1 to 3 V/cm through transducer arrays placed on the patient's skin close to the tumor. All TTFields studies predicted variability in treatment response among patients, whereas in vitro experiments indicate that the magnitude and direction of the electric field in the tumor might be crucial determinants of efficacy. Differences in the field might arise from varying tumor positions or array placement. By investigating different scenarios within a realistic human head model we hope to advance our understanding of TTFields therapy in clinical practice. We constructed a model from MRI data to calculate the electric field distribution in the brain using the Finite Element Method. An anisotropic electrical conductivity tensor was estimated using diffusion tensor imaging data. The head model contained different tissue types: scalp, skull, cerebrospinal fluid, gray and white matter. Additionally a virtual spherical tumor was included, two positions for the tumor were considered. Transducer arrays were placed on the scalp to model the commonly used device for TTFields delivery. One additional setup of the two transducer pairs was specifically adapted to the second tumor position. The results predict that the electric field strength exceeds the assumed therapeutic threshold value of 1 V/cm in both tumors for both active array pairs. For the second tumor the adapted transducer layout improved field delivery. The average field strength in the tumor further depends on tumor electrical properties. Yet a cystic and a solid tumor experience the same average field strength when treated with TTFields. As a next step towards personalized TTFields therapy, we will explore possible benefits of individualized treatment planning.