September, 2011Accurate estimations of age at death are essential to law enforcement in the identification of human skeletal remains, but in complicated cases decomposition and other factors can create uncertainty in the estimate.
In those cases, forensic anthropologists must employ more than one scientific method for determining age at death, each of which may yield a different estimate. Until now, there has been no standardized way to combine the results of several methods to arrive at a single, confident estimate.
Under a $ $417,595 grant from the National Institutes of Justice, forensic anthropologist Danny Wescott of Texas State University-San Marcos will spend the next two years developing a standardized, easy-to-use method for determining age at death based on multiple criteria. Wescott is director of the Forensic Anthropology Center at Texas State (FACTS), which oversees the largest forensic anthropology research and training facility in the United States. Research done at FACTS and at its 26-acre outdoor lab by faculty and graduate students provides new information on the processes of human decomposition—information that assists law enforcement in identifying human bodies and establishing the time and nature of death.
Until now, forensic anthropologists typically have determined age at death by using one or more of the methods developed for collecting information from different regions of the skeleton known to change in predictable ways as a person ages. None of the methods is capable of producing an accurate estimate in every case, however, because of genetic variation, environmental factors, diet, disease, activity levels and the decomposition process. As Wescott explained, forensic anthropologists need a “multi-factorial method” that combines information from different parts of the skeleton and that accounts for inaccuracy in each of the methods, to produce a single, accurate estimate of age at death. Currently, no standardized guidelines exist for taking variables into consideration, and forensic anthropologists often have to rely on their personal experience, rather than on statistical information, to produce an estimate of age at death.
“A number of variables create uncertainty in age-at-death estimations,” Wescott said. “These variables can include the quality of bone, the appropriateness of the investigative method for the target age group, and observer error. Currently, there are no ‘best practice’ guidelines in forensic anthropology for combining the results of multiple age-determining methods for a single skeleton, to produce a confident estimate. What we need is a standardized method that can produce an accurate estimation of age-at-death in every case.”
Wescott and three colleagues from Mississippi State University and the University of Missouri recently produced a new method for estimating age at death, which Wescott says has several advantages over current methods.
“Our method allows investigators to use nearly any well-established age-at-death indicator method and to fuse information about the accuracy of the method with the other variables that can cause uncertainty, to produce an accurate estimate for a single skeleton,” Wescott said.
Wescott’s method uses the Sugeno Fuzzy Integral—a mathematical tool—to produce confidence in age-at-death estimates. Fuzzy logic deals with reasoning that is approximate rather than exact. In contrast with traditional logic in which information is either true or false, fuzzy logic variables can be partially true. In Wescott’s method, truth values are established for each set of skeletal information so that, when the information sets are combined, they produce a confident outcome.
“Fuzzy logic can be used to make a complex decision when there are multiple sources of information available,” Wescott said. “Our method makes the process explicit for analyzing qualitative fuzzy sets, so that forensic anthropologists, law enforcement agents, lawyers, and other members of the medico-legal community can understand how to interpret the results.
“No other method allows for the fusion of so many kinds of skeletal information,” Wescott continued. “Our method can be used for both adult and immature skeletons, it can be customized to meet the investigator’s needs on specific cases, and it provides informative graphs and a standardized reproducible way to generate linguistic descriptions of age-at-death estimations.”
Under the NIJ grant, Wescott and his colleagues from Mississippi State University and the University of Missouri will develop a program that will allow forensic anthropologists to submit multiple indicators of age, bone quality information, and other factors, to produce a confident age-at-death estimate in graphic, numeric, and linguistic form. The NIJ grant will also provide funding for two graduate students and at least one undergraduate student at Texas State, who will assist in standardizing the procedure for general use.