Towards ultrasound hyperthermia safe treatments using computational intelligence techniques

Abstract

A key feature for safe application of hyperthermia treatments is the efficient delimitation of the treatment region avoiding collateral damages. The efficacy of treatment depends on an ultrasound power intensity profile to accomplish the temperature clinically required. Many hyperthermia procedures proposed in the literature rely on a-priori knowledge of the physical properties of tissue. The soft computing models presented in this article are only based on measured data, collected from tissue phantoms reflecting the reactions of human tissues to ultrasounds. From homogeneous to heterogeneous tissues, different soft computing techniques were developed accordingly to experimental constraints. The present state of development is nearly approaching the identification of a computational model to be safety applied in in-vivo hyperthermia sessions.