Variability of dental cone beam CT grey values for density estimations

Ruben Pauwels, O. Nackaerts, N. Bellaiche, H. Stamatakis, K. Tsiklakis, A. Walker, H. Bosmans, R. Bogaerts, R. Jacobs, K. Horner

    Research output: Contribution to journalArticlepeer-review


    Objective: The aim of this study was to investigate the use of dental cone beam CT (CBCT) grey values for density estimations by calculating the correlation with multislice CT (MSCT) values and the grey value error after recalibration. Methods: A polymethyl methacrylate (PMMA) phantom was developed containing inserts of different density: air, PMMA, hydroxyapatite (HA) 50mgcm-3, HA 100, HA 200 and aluminium. The phantom was scanned on 13 CBCT devices and 1 MSCT device. Correlation between CBCT grey values and CT numbers was calculated, and the average error of the CBCT values was estimated in the medium-density range after recalibration. Results: Pearson correlation coefficients ranged between 0.7014 and 0.9996 in the full-density range and between 0.5620 and 0.9991 in the medium-density range. The average error of CBCT voxel values in the medium-density range was between 35 and 1562. Conclusion: Even though most CBCT devices showed a good overall correlation with CT numbers, large errors can be seen when using the grey values in a quantitative way. Although it could be possible to obtain pseudo-Hounsfield units from certain CBCTs, alternative methods of assessing bone tissue should be further investigated. Advances in knowledge: The suitability of dental CBCT for density estimations was assessed, involving a large number of devices and protocols. The possibility for grey value calibration was thoroughly investigated. © 2013 The British Institute of Radiology.
    Original languageEnglish
    Article number20120135
    JournalBritish Journal of Radiology
    Issue number1021
    Publication statusPublished - 1 Jan 2013


    Dive into the research topics of 'Variability of dental cone beam CT grey values for density estimations'. Together they form a unique fingerprint.

    Cite this