For the first time, science-based standards have been established for identifying human remains using X-rays.

A team of researchers, led by Dr. Ann Ross, lead author of a paper on the new standards and a professor of anthropology at North Carolina State University, set out to evaluate the use of various anatomical features most frequently X-rayed in an attempt to make positive identifications of bodies through radiographic comparison.

The team analyzed and compared antemortem and postmortem radiographs of the lateral craniofacial — the side of the skull — and the spine and femur. Twenty individuals’ lateral craniofacial X-rays were used, 50 individuals for the spine, and 23 individuals for the femur, since these regions are among the most frequently X-rayed in a clinical setting.
“Our intention is to go through different areas and develop standards. What happens now is that either a forensic anthropologist or medical examiner will take the antemortem X-ray and the one taken at autopsy and compare them side to side,” Ross told HCB News.

She added that there have been no standards as to what is the minimum number of concordant points necessary to be comfortable with a positive identification. In this case, the concordance will be the features that are found in both images of the X-ray.

Following the evaluations of the standards, the team then used unmatched X-rays where the remains have not been identified to be the same person, to test the accuracy of the standards themselves. One way they tested them was to see how likely the standards were to accurately identify a body. Another was to run separate analyses to see how likely the standards were to misclassify an identification.

The standards were found to vary significantly. Using as few as two specific traits, a side X-ray of the skull could identify a body with a certainty of 97 percent and a 10 percent misclassification, as long as there were no inconsistencies in the shapes of the skull X-rays, according to the study.

As for the vertebrae, the standards could provide more than 98 percent accuracy and a 7 percent misclassification based on a single trait. However, for the lower back vertebrae, there was a 40 percent misclassification, even if four matching traits were identified.

The goal of this study, according to Ross, is to bring standards and consistency to radiographic ID, along with quantification to the process, so identification is not based on one expert’s individual opinion.

“I hope this is moving us forward toward a more standard protocol for our forensic cases, which has been lacking for a number of years. I hope people will take this and go and start gathering X-rays and conducting their own studies on different parts,” Ross said.

Back to HCB News

Population Health Homepage