Angular dispersive X-ray diffraction signatures from human liver tissues using synchrotron radiation

Full text: Amorphous materials such as human tissue display an X-ray diffraction pattern that is a representation of the molecular structure of the material. The diffraction patterns are due to constructive interference of waves scattered from electrons within the same atom, molecule and neighbouring molecules. Human tissue has short range order that allows characteristics of diffraction to be seen and differences between tissue types may be identified. This study utilises the X-ray signatures of human liver tissues as a method of differentiating between two liver tissue classifications, i.e. cancerous and normal tissue. X-ray diffraction has been used by other studies as a method of breast tissue classification but to the best of the authors' knowledge, this paper is the first to present X-ray diffraction signatures of cancerous and non cancerous liver tissues. This study presents the X-ray diffraction profiles of 30 matched pairs of human liver samples. The cancer samples are taken from areas of the liver known to have secondary colorectal metastases and the normal tissue is taken from the same patient but at a distance from the tumour area. Angular dispersive X-ray diffraction was used to measure the distribution of the coherently scattered photons utilising synchrotron radiation. The experiment took place at the European Synchrotron Radiation Facility (Grenoble, France). The measurements were made for an angular range of 5 deg to 50 deg, an incident energy of 13keV and a beam area of 1 x 1 mm2. This corresponds to a momentum transfer range of approx. 0.5 nm-1 to 4.4 nm-1. Following the measurements, the samples were analysed for histology by an experienced histopathologist. The measured profiles are corrected for the synchrotron flux and self-attenuation. The diffraction spectra are pre-processed (smooth and background subtraction) and deconvoluted into a number of Gaussian peaks using PeakFit software. Fitting parameters such as peak amplitude and area are extracted and used for statistical analysis. We will show that there are clear differences between the normal and diseased tissue and these results are presented and discussed