Towards 3D Luminescence Thermometry: an approach based on lanthanide nanoparticles and dimensionality reduction techniques

Abstract

One of the main problems for the application in biomedicine of spectral-based luminescent nanothermometers is the distortion of the shape of the emission spectra due to the presence of an organism’s tissues. Different solutions have been proposed to overcome this problem, such as exploiting the temperature dependence of the lifetime of nanoparticles rather than their emission, because lifetime is not affected by the presence of biological tissues. This solution works very well especially in the case of nanoparticles doped with lanthanide ions, since they have longer lifetimes in the order of hundreds of microseconds. The aim of this project was to explore the possibility to take advantage of these tissue-induced spectral changes. Rather than their application in thermometry, it was investigated first their reliability in the prediction of the thickness of a biological tissue, and later, they were exploited to build-up the case for 3D nanothermometry. Nanoparticles doped with lanthanide ions (NaYF4@NaYF4:Nd,Yb@CaF2) were used as luminescent nanothermometers, knowing that their lifetime was dependent on temperature, while the features of their emission spectra were dependent on the thickness of the tissue. Hyperspectral images were collected at different temperatures and then dimensionality reduction techniques (specifically Principal Component Analysis) were applied to better visualize the dependence of emission spectral changes with the thickness of the tissue. This analysis led to a new representation of the system that was finally used to make the regression for the 3D image reconstruction of a simple conic-shaped tissue.