Correlation of X-ray diffraction signatures of breast tissue and their histopathological classification

Abstract

This pilot study examines the correlation of X-ray diffraction (XRD) measurements with the histopathological analysis of breast tissue. Eight breast cancer samples were investigated. Each sample contained a mixture of normal and cancerous tissues. In total, 522 separate XRD measurements were made at different locations across the samples (8 in total). The resulting XRD spectra were subjected to principal component analysis (PCA) in order to determine if there were any distinguishing features that could be used to identify different tissue components. 99.0% of the variation between the spectra were described by the first two principal components (PC). Comparing the location of points in PC space with the classification determined by histopathology indicated correlation between the shape/magnitude of the XRD spectra and the tissue type. These results are encouraging and suggest that XRD could be used for the intraoperative or postoperative classification of bulk tissue samples.

Determination of ingredients in packaged pharmaceutical tablets by energy dispersive X‐ray diffraction and maximum likelihood principal component analysis multivariate curve resolution‐alternating least squares with correlation constraint

“…To account for matrix effects, separate internal calibration models using the correlation constraint are applied using the calibration samples within each subset. The two local models are defined as follows: where c * corresponds to the correctly scaled reference concentrations and c to the concentrations resolved by the least squares step 5 . Calibration coefficients b are calculated using the resolved concentrations and reference concentrations in the calibration sets for the two subsets: …”

“…where c * corresponds to the correctly scaled reference concentrations and c to the concentrations resolved by the least squares step. 5 Calibration coefficients b are calculated using the resolved concentrations and reference concentrations in the calibration sets for the two subsets:…”

“…Polycrystalline materials are of particular interest in screening contexts, where investigations have shown the potential of EDXRD in identifying the composition of solid‐dosage pharmaceutical formulations, powder‐form illicit drugs and plastic explosive materials 1–4 . Furthermore, the technology has shown potential in diagnostic medicine to discriminate between healthy and cancerous tissue in breast cancer screening 5 …”

“…[1][2][3][4] Furthermore, the technology has shown potential in diagnostic medicine to discriminate between healthy and cancerous tissue in breast cancer screening. 5 Traditional application of X-ray diffraction in a laboratory setting, either with a synchrotron radiation source or a conventional mono-energetic angular dispersive X-ray diffraction (ADXRD) system, enables high-resolution sample profiles to be recorded. The profiles have high spectral selectivity and can be parametrically fitted and then crossreferenced against databases in order to characterise and identify the samples.…”

A review of the applications of Raman spectroscopy for breast cancer tissue diagnostic and their histopathological classification of epithelial to mesenchymal transition

“…Concomitant with the latter, nanoscale changes in the structure associated with disease can also be used to discriminate between different tissue types, highlighting the importance of the tumour‐associated stroma in breast cancer initiation and progression. The main component of such stroma is abnormal collagen, the orientation and amount of which can directly influence epithelial growth and transition, including response to neoadjuvant therapy . The mechanism is similar to the epithelial to mesenchymal transition (EMT) process that has been characterised in embryonic development .…”