Inverse optimization based non-invasion technique for multiple tumor detection in brain tissue.

The aim of this article is to predict multiple tumors via inverse analysis, which are accumulated within a human brain tissue utilizing temperature profiles available easily through infrared thermography. Thermophysical characteristics like blood perfusion rate, which differs in malignant and normal tissue was also retrieved. Pennes bioheat model was used for the study of heat transfer in human body. Considering brain tissue, scalp and skull bone, a 1-D study was made. For the forward analysis, finite element approach was utilized, where temperature distributions were obtained by assuming presence of two tumors. The adopted forward method was validated theoretically and experimentally and the results were found satisfactory. Blood perfusion rates were determined using an inverse optimization algorithm, namely, differential evolution, which clearly depicts the presence of tumor for the specified retrieved value. Furthermore, the maximum deviation of temperature field obtained from the estimated value of blood perfusion rate from exact ones was found to be 0.008%. [ABSTRACT FROM AUTHOR]