Deriving Hounsfield units using grey scale of a CBCT?: author response

Abstract

We appreciate the reader's interest in our research paper "Deriving Hounsfield units using grey levels in cone beam computed tomography" in the September 2010 issue of DMFR. 1 We have subsequently applied this method in a clinical situation and the results will be submitted shortly for publication by the same authors. This should help to explain some of the issues raised in this letter to the editor.We are aware of the paper by Bryant J A et al 2 and we will attempt to address some of the issues regarding non-uniformity in the flat panel detector and the issue of noise raised in this paper in a future article.At this time we are unable to scan the device in an i-Cat CBCT machine as we do not have one of these scanners in our institution. We have several other CBCT devices but not this particular manufacturer's device.

Deriving Hounsfield units using grey levels in cone beam CT: a clinical application

“…While a CBCT scan is also useful in determining bone quantity, bone height and proximity of adjacent structures, it lacks the ability to quantify bone quality in a meaningful manner. This study has demonstrated a clinical application of a method developed by Mah et al 9 to derive Hounsfield units using grey levels in the CBCT volume. The application of this method to derive Hounsfield units can provide a standardized method to assess bone quality similar to that found in medical CT.…”

“…While the original grey level values are not realistic Hounsfield units it is at least possible to distinguish between the two kinds of bone. Figure 5 shows the distribution of ''corrected'' grey levels using the method of Mah et al 9 While the average values for the two materials are now realistic Hounsfield units, the variability of effective energies results in such a tremendous overlap between the values of the two materials that it is difficult to distinguish one from the other. This variability would seem to be a serious deficiency in this method of calibration.…”

“…Given that there are many CBCT machines currently being used in dentistry which operate with differing tube potential and filtration as well as reconstruction algorithms, it is not really surprising that the grey levels obtained on dental CBCT scanners vary widely. 9 As Hounsfield units are routinely used in CT to compare densities of different materials, it would seem that the adoption of Hounsfield units would also offer a method of comparing the relative densities of tissues in a CBCT scan. 10 In a recent study, Mah et al 9 introduced a method where Hounsfield units could be derived using grey levels in CBCT.…”

“…9 As Hounsfield units are routinely used in CT to compare densities of different materials, it would seem that the adoption of Hounsfield units would also offer a method of comparing the relative densities of tissues in a CBCT scan. 10 In a recent study, Mah et al 9 introduced a method where Hounsfield units could be derived using grey levels in CBCT. This was an extensive study using 11 different CBCT scanners in 3 different situations: (1) phantom alone; (2) phantom in small water container to mimic a small patient, such as a child; and (3) phantom in a large water container to mimic an adult patient head.…”

“…It was concluded that the method was effective in deriving Hounsfield units using grey levels within approximately the same variability range as that observed between medical CT scanners, which was less than 6.4%. 9 The purpose of the present study was to apply this in vitro method to derive Hounsfield units from CBCT grey levels in a clinical situation with human patients.…”