Date: Mon 5.11 (TS335) - Tue 6.11 (TS287)
Time: 10:00 - 14:00 (both days)
Due to the complex and inhomogeneous structure of biological tissues, the analysis of imaging data collected with various optical biopsy methods is often complicated and time consuming. The major challenge here is to understand the peculiarities of light propagation and link it with advanced image/data classification pipelines.
This short course will be dedicated to the exploration of the potential of the emerging Artificial Intelligence (AI) based methods to the inverse problem of light transport in scattering media such as human skin and consists of three major parts. 1) Creation of voxelized, biophysically-based optical model of the tissue that considers spatial/volumetric variations in both structural e.g. surface roughness and chromophore concentration changes in skin layers for different ethnic and age groups such as distribution of blood, melanin, collagen, index of blood oxygen saturation, water, pigment content, etc. 2) The development and validation of a hyperspectral Monte-Carlo (HyperMC) based approach for forward simulation of spectra and Bidirectional Scattering-Surface Reflectance Distribution Function (BSSRDF) 3) Building a spectral image classification pipeline based on Artificial Neural Networks (ANNs) by implementing and training several configurations of ANNs classifiers that fit for the scattering and absorption properties of the tissues.
Computer simulation and training are accelerated by parallel computing on Graphics Processing Units (GPUs) using Compute Unified Device Architecture (CUDA) and Cloud-based environment. Open-source machine learning frameworks (e.g. Tensorflow) are used to measure and validate each ANN’s performance. The results chromophore mappings for parameters such as distributions of melanin, blood vessels, oxygenation, simulation of BSSRDFs, reflectance spectra of human tissues, corresponding colors and 3D rendering examples of human skin appearance will be presented and compared with in vivo experimental data obtained during clinical studies.
Last updated: 18.10.2018