Abstract
BACKGROUND: At present, infrared (IR) imaging is used both as a non-invasive and a non-ionizing technology. Using an IR camera, it is possible to measure body surface temperature in order to detect tumors and malignant cells. Tumors have a high amount of vasculature and an enhanced metabolism rate, which may result in an increase in body surface temperature by several degrees above its normal level. METHODS: Using thermograms, it is possible to assess various tumor parameters, such as depth, intensity, and radius. Also, by solving for Penne's bioheat equation, it is possible to develop the analytical method to solve for inverse heat conduction problem (IHCP). RESULTS: In the present study, these parameters were optimized using artificial neural networks in order to localize the heat source in the medium (i.e. female breast) more accurately. CONCLUSION: Eventually, a new formula was derived from Penne's bioheat equation to estimate the depth and radius of the embedded heat source. Moreover, by analyzing the data, errors of the parameters could be estimated.