Hyperspectral imaging as a technique to evaluate peritonitis in mouse models

Hyperspectral imaging (HSI) is an optical technique that combines imaging and spectroscopy. Images can be processed spectroscopically, extracting information about tissue biochemistry and morphology that can be of importance to physicians.

Peritonitis is an inflammation of the peritoneal cavity that can result in death. In animal model studies of the disease, golden standard is histology, while biomolecular techniques can also be used. These techniques only provide either good spatial resolution or area coverage. A technique with large area coverage and resolution is thus desired.

We present a classification method for determination of the state of peritonitis using HSI. The technique provides both good spatial resolution and coverage of the whole organ. Spectra are acquired in reflection geometry and processed using a custom model of light transport in thin samples. Tissue parameters including oxygenation and scattering properties are extracted from the images and used to assess disease state. Per subject scores are used to successfully separate control from model groups. HSI obtained scores are also compared with existing techniques (histology, PCR and Western blot) showing good concordance, meaning that HSI can be a useful tool for disease evaluation.