CT imaging vs. traditional radiographic imaging for evaluating Harris Lines in tibiae

This paper is the first to systematically investigate computer tomography (CT) images vs. ordinary flat plane radiography for evaluating Harris Lines (HL) on tibiae. Harris Lines are traditionally investigated using radiographic images and recorded as either present or absent, or by counting the number of HL. Seventy-four pairs of human sub-adult and adult archaeological tibiae were used in this study. Both image methods were tested for intra- and inter-observer agreement and the methods were then compared. Analysis was performed with the tibiae divided into younger (n = 19) and older sub-adults (n = 26) and adults (n = 29), as well as all tibiae combined (n = 74). This study found that the intra- and inter-observer agreement was very similar for each method, but both image methods performed less well for counting the number of HL. Direct comparison between CT images and radiographic images showed good agreement for presence or absence of HL and less agreement for counting HL. There appeared to be no systematic pattern for dividing the material according to bone length (and thereby age). Overall, this paper concludes that the performance of CT images is comparable to traditional flat plane radiography for evaluating HL on archaeological dry bone specimens of tibiae. Further, due to the lower performance of counting the number of HL, it appears advisable to only record the presence or absence of HL rather than counting the number of HL.