Artificial neural networks trained on simulated multispectral data for real-time imaging of skin microcirculatory blood oxygen saturation.

Abstract

Imaging blood oxygen saturation ( ) in the skin can be of clinical value when studying ischemic tissue. Emerging multispectral snapshot cameras enable real-time imaging but are limited by slow analysis when using inverse Monte Carlo (MC), the gold standard for analyzing multispectral data. Using artificial neural networks (ANNs) facilitates a significantly faster analysis but requires a large amount of high-quality training data from a wide range of tissue types for a precise estimation of . We aim to develop a framework for training ANNs that estimates in real time from multispectral data with a precision comparable to inverse MC. ANNs are trained using synthetic data from a model that includes MC simulations of light propagation in tissue and hardware characteristics. The model includes physiologically relevant variations in optical properties, unique sensor characteristics, variations in illumination spectrum, and detector noise. This approach enables a rapid way of generating high-quality training data that covers different tissue types and skin pigmentation. The ANN implementation analyzes an image in 0.11s, which is at least 10,000 times faster than inverse MC. The hardware modeling is significantly improved by an in-house calibration of the sensor spectral response. An in-vivo example shows that inverse MC and ANN give almost identical values with a mean absolute deviation of 1.3%-units. ANN can replace inverse MC and enable real-time imaging of microcirculatory in the skin if detailed and precise modeling of both tissue and hardware is used when generating training data.