Utilization of Multi-spectral Images in Photodynamic Diagnosis

This paper introduces multi-spectral images for healthy and cancerous parts of human skin. It compares light spectrum calculated from those images with spectrum obtained from simulation. First the mathematical model of tissue and Monte Carlo algorithm of light propagation in turbid media is presented. This theory was then extended to imitate the fluorescence phenomenon, necessary for cancer recognition. Then the processing method of non-normalized multi-spectral images was described. Finally both results were compared to confirm that the assumed model is correct. Having all those information it will be possible to simulate such environment, which applied into reality, would make the cancer diagnosis much faster.