Multispectral fibre imaging of lung disease

Respiratory diseases are prevalent throughout the world and include a range of conditions such as pneumonia, tuberculosis, and many others. In the clinic, we see there is an unmet need for quick and accurate diagnosis of lung infection and inflammation in intensive care.
The Proteus project (www.proteus.ac.uk) aims to improve the detection and diagnosis of pulmonary infection and inflammation by employing targeted fluorescent molecules (Smartprobes) to visualise bacterial burdens in the distal lung. Sometimes the signal from the Smartprobes is difficult to detect on a background of lung tissue, which is brightly fluorescent due to the abundance of collagen and elastin. In addition to resolving probes from intrinsic fluorescence, multiple probes may have overlapping emission spectra. This is particularly true for many well-established fluorophores that reside in the green region of the spectrum. If imaging with fluorophores that have distinct emission spectra is not possible or desirable, then spectral sorting of the signals can be carried out.
Here we present a simple coherent fibre bundle based[1] widefield ratiometric imaging system consisting of a single colour LED illumination source (480nm) that is capable of enhancing the contrast between similar spectra of pathologically relevant fluorescence signals without complex spectral unmixing. To visualise disease, Smartprobes are delivered into the lung through capillary channels down a novel, cheap, and disposable fibre bundle developed by the University of Bath. The bundle features an imaging fibre which consist of 8100 cores with a 450µm corner to corner field of view
Contrast enhancement is carried out on dual images of the same field of view from different parts of the spectrum by post processing of the images. This enables us to interpret better the images produced both in autofluorescence and molecular imaging contexts.
[1] A Perperidis, H Parker et al, Optics Express 25 (10), 11932, (2017)