|Institution:||Kansas State University|
|Department:||Department of Electrical and Computer Engineering|
|Keywords:||Ablation; Microwaves; Electrical Engineering (0544)|
|Full text PDF:||http://hdl.handle.net/2097/19058|
Microwave ablation is a minimally invasive therapeutic modality used for the treatment of cancer in various organs. In this procedure, microwave energy is sent through a thin antenna placed inside the tumor. The microwave energy radiated from the antenna generates heat which kills the tumor cells by necrosis. During multiple-applicator microwave ablation, geometric estimates of treatment outcome are typically obtained by assuming parallel insertion of the applicators. This assumption is based on the guidelines provided in the brochures of antenna manufacturing companies. This assumption is flawed because it is rare to insert the antennas in parallel configuration due to the flexible nature of the antennas and the presence of intervening organs. Furthermore, movement of patients during the treatment procedure alters the position of the antennas. In order to see the effect of non-parallel insertion of antennas, model-based treatment planning may be instructive. Treatment planning can also determine the changes needed to be made for prospective ablation therapy if the antennas are not positioned in their ideal parallel configuration. This thesis provides a detailed computational comparison of the skewed configurations of microwave antennas to their closest parallel configurations. The metric used for com-paring the similarity between the cases is Dice Similarity Coefficient (DSC). Experimental results to validate the computational data are also discussed. Computations were done by using realistic cases of antenna positions obtained from Rhode Island Hospital.