2022

Annala, Leevi; Pölönen, Ilkka

In the medical field, there is a need for non-invasive diagnostic tools. One particular research area is skin cancer diagnostics. Here, we study Kubelka–Munk model and stochastic skin reflectance model, which we combined from two sources to better reflect the physical structure of the skin. Our objective is to compare the models to each other in terms of accuracy, usefulness, and biophysical parameter retrieval using a convolutional neural network. The results are promising. Both models are found suitable options for further research and used Stochastic Model similar to Kubelka–Munk in terms of accuracy. In physical parameter retrieval, both models perform moderately. Inverted models reasonably retrieve the pigment concentrations from the simulated test data set. With empirical testing data, the inverted models are mutually consistent.