PLoS ONE (Jan 2018)

Novel radiomics evaluation of bone formation utilizing multimodal (SPECT/X-ray CT) in vivo imaging.

  • Ferenc Budán,
  • Krisztián Szigeti,
  • Miklós Weszl,
  • Ildikó Horváth,
  • Erika Balogh,
  • Reem Kanaan,
  • Károly Berényi,
  • Zsombor Lacza,
  • Domokos Máthé,
  • Zoltán Gyöngyi

DOI
https://doi.org/10.1371/journal.pone.0204423
Journal volume & issue
Vol. 13, no. 9
p. e0204423

Abstract

Read online

Although an extensive research is being undertaken, the ideal bone graft and evaluation method of the bone formation draw still a warranted attention. The purpose of this study was to develop a novel multimodal radiomics evaluation method, utilizing X-ray computed tomography (CT) and single photon emission computed tomography (SPECT) with Tc-99m-Methyl diphosphonate (Tc-99m-MDP) tracer. These modalities are intended to provide quantitative data concerning the mineral bone density (after evaluation it is referred to as opacity) and the osteoblast activity, at the same time. The properties of bone formation process within poly (methyl methacrylate)-based bone cement graft (PMMA) was compared to that of albumin coated, sterilized, antigen-extracted freeze-dried human bone grafts (HLBC), in caudal vertebrae (C5) of rats. The animals were scanned at 3 and 8 weeks after surgery. In both groups, the mean opacity increased, while the mean Tc-99m-MDP activity decreased. The later parameter was significant (n = 4, p = 0.002) only in HLBC group. The linear regression analysis of PMMA-treated group variables (mean opacity increase; mean Tc-99m-MDP activity decrease), revealed a negative correlation with the medium strength (r = 0.395, p = 0.605). Whereas, it showed strong positive correlation when HLBC group variables were analyzed (r = 0.772, p = 0.012). These results indicate that using HLBC grafts is advantageous in terms of the osteoblast activity and bone vascularization over PMMA cement. Using this regression analysis method, we were able to distinguish characteristics that otherwise could not be distinguished by a regular data analysis. Hence, we propose utilizing this novel method in preclinical tests, and in clinical monitoring of bone healing, in order to improve diagnosis of bone-related diseases.