Disease-related spectral regions attributed to glycated products were captured through near-infrared spectroscopy and when paired with machine learning classification models, produced high levels of predictive accuracy.
Wilson et al. (Thu,) studied this question.