Investigating the Feasibility of Using HEXITECMHZ and Fully-Spectral X-ray Imaging to Detect and Diagnose Breast Tumours: an In Silico Study

Photon counting (PC) detectors have drawn a great deal of attention due to their ability to leverage spectral information from a polychromatic x-ray beam to improve soft tissue contrast and material discrimination in medical applications. They are nonetheless limited to sorting the entire polychromatic beam into a relatively small number of energy bins, which can span many 10s of keV each. In contrast, fully-spectral x-ray (FX) detectors produce full energy spectra for each pixel in the image, allowing for far more complex analysis of the materials under investigation. This work sets out to determine if the increased spectral information provided by FX detectors can be used to detect differences between breast tumours and healthy breast tissue by analysing their trace element compositions. An existing PC simulation framework (called CoGI) was modified to model the HEXITEC
MHZ
FX detector developed by Rutherford Appleton Laboratory. This detector was simulated, along with the x-ray fluorescence spectra associated with both healthy and cancerous breast tissue phantoms. Of the trace elements considered, only 4 were detectable (Fe, Zn, Br and Rb) and of these the ratio between Fe and Zn provided the strongest method for differentiating the phantoms.