Blood will tell

Hank Hogan, hank@hankhogan.com

The job of criminal investigators may have gotten a bit easier, thanks to a method based on Raman spectroscopy that quickly determines the species of origin of a dried blood sample. With this knowledge, police detectives can ensure that the right samples are collected at a crime scene.

In developing the method, associate chemistry professor Igor K. Lednev and graduate student Kelly Virkler of the University at Albany, State University of New York, had to overcome a basic problem. For simple compounds, the standard approach is to compare collected Raman data against a library of known spectral signatures. “We cannot do this for complex plex species like traces of blood,” Lednev said.

One reason for this, he explained, has to do with the inhomogeneity of dried samples. A second is the variation among individuals within the same species. To get around these problems, the researchers collected the near-infrared Raman spectrum at 16 points from each dried blood sample. They had eight samples from each of three species: human, cat and dog. After averaging the 16 results for each of the 24 individual samples, they used statistical methods to extract the linear factors behind the spectroscopic data.

This graphic illustrates how the dried blood of each species has its own distinct nonoverlapping Raman spectrum, as represented at the bottom of the image. Courtesy of Aliaksandra Sikirzhytskaya and Igor Lednev, University at Albany, State University of New York.

They found that as few as three components explained most of the variation, with the relative contribution of the components different for the three species. These three components together were distinct enough that the researchers could say, with 99 percent certainty, whether a sample came from a human, a cat or a dog.

While these findings are encouraging, more work must be done before the technique can be used in the field, Lednev said. One issue is cost. The researchers used a general purpose scientific device, a Renishaw confocal Raman microscope with a 785-nm excitation laser. Police departments require a much less expensive setup, which might be achieved with a specialized instrument.

A second issue is false positives – samples identified as being human that in fact are not. That rate must be zero, according to Lednev. He noted that achieving this standard is possible, although the result might be more cases where no identification is made.

A final issue is that forensic professionals usually are not Raman experts. This is being kept in mind as the technique is being turned into a methodology and tools suitable for field use are being developed. “We need to make this software absolutely automatic,” Lednev said.

Eventually, this approach could lead to other fast nondestructive tests. They might reveal whether a blood sample came from a man or woman, from a smoker or nonsmoker, or from someone with certain blood-borne conditions or diseases. These techniques could also be applied to the battlefield.

The group has just secured funding from the National Institute of Justice for further research and development of the technique. On the photonics side, Lednev noted that collecting the Raman spectrum further into the infrared would help because it would eliminate sample fluorescence.