Infrared camera system could detect blood at crime scenes

Crime scene investigation could get a bit less messy. Researchers from the University of South Carolina have developed an infrared camera system that can detect blood -- and the new technology could one day replace the chemicals currently used.

Crime scene investigation could get a bit less messy. Researchers from the University of South Carolina have developed an infrared camera system that can detect blood -- and the new technology could one day replace the chemicals currently used.

Here are excerpts from my recent interview with Stephen Morgan, a chemistry and biochemistry professor who collaborated on the work with professor Michael Myrick and students:

Why is there a need for a better type of technology to analyze blood at crime scenes?

Analyzing blood at crime scenes is a three-step process. The first step is a presumptive detection technique. That's a quick technique used to identify the possible presence of blood. The types of presumptive tests are usually simple chemical reaction reagents that give a positive reaction to blood. But there are issues with chemical reagents. These tests can suffer from false positive and false negative results. They're not very specific. There are some common household materials that react with [these reagents], including the pulps of certain root vegetables, copper metal and spray paint.

The problem with a false positive is the investigator has to test that spot because there's an indication there might be blood there, so time and resources are wasted. A false negative result is more serious because it means blood is there, but you're not detecting it. After the presumptive test, you do a confirmatory test. There are various levels of these. The best can confirm pretty well, but these tests are more involved and cannot be done at the site very easily. Each of these higher levels requires more and more work. If you've got a false positive sample that doesn't really contain blood, that's a lot of wasted time.

What's the purpose of identifying the blood? The crime scene shows like to talk about the blood spatter patterns -- from the direction and shape of a blood drop you can potentially find out something about what occurred. The real purpose of blood detection these days is to extract DNA from the blood and use that DNA to potentially identify a suspect. One of the problems that I point out is it's counterintuitive to contaminate the whole crime scene with a chemical that may or may not give you good statistics on detection if you're trying to recover DNA.

But your system doesn't use chemicals.

Our concept was to shine a light on the crime scene and light will interact with chemicals on surfaces. We're using infrared light. You can't see it with the human eye. Some of the light reflects back, as if from a mirror. What we're interested in is the light that penetrates a little bit into the surface. That happens with all substances. That light interacts with the surface and some of it may be absorbed. When it is readmitted by the substance, it will contain a pattern or spectrum that is related to the chemical nature of that surface. The amount of infrared light reflected off the surfaces in a normal room is like 1,000 light bulbs, but they're part of the spectrum we don't see. When we shine an extra infrared light on a surface in that room, we're looking for a minute portion of light that's reflected back. This is an extraordinarily difficult measurement to make.

When we first set this up, we were using a commercially-available infrared camera and we were looking at blood. We needed to design a more sensitive way to detect that light. We did it by flashing the light on and off. We subtract the signal when the light is off from the signal when the light is on. That extra increment of light is the light being reflected. That's the simple explanation of it. The technique we're using is a "lock-in amplifier approach." When we were able to do that, we knew we had a device that worked.

How far along is the work now?

We're funded by the National Institute of Justice. This is the premier forensic research funding institution in the United States. We responded to a call for proposals in 2007. That project ended last year and we published a number of papers. This instrument is on a bench on our laboratory. It's never been removed from that laboratory. We've done some simple testing of our detection limits and ability to detect blood on different types of surfaces. We've done other types of studies, as well. Our preliminary estimate is that we can detect blood down to 200 times diluted. Sometimes the perpetrator will try to clean up the crime scene by washing it. If the blood is diluted, it becomes difficult to see visually. Our technique seems pretty sensitive.

We have submitted a request for renewal of our funding. In that proposal are plans to collaborate with the premier forensic lab in South Carolina. We're going to collaborate with them in real crime scene situations. We're going to place an instrument in their evidence room at their lab and it will be used specifically for detection of blood. Another part of this proposal is to collaborate with a company that might be interested in commercializing this idea. From the beginning, our goal was to produce research that would benefit investigators. The advantage of this technique is that we're not destructive and we don't contaminate the crime scene.

The next phase of the project will be to build more portable instruments. We imagine a handheld device that can be positioned on a tripod. There will be a laptop computer attached to it. It can be set up in a crime scene. The ability to make it portable is the next stage and that would be one advantage of commercialization.

This post was originally published on