An international team of engineers led by KU Leuven University in Belgium have built an artificial intelligence (AI) algorithm that scans facial image databases and identifies individuals likely to match with DNA recovered from crime scenes.

The shape of the human face is largely determined by thousands of different genes, so researchers believe that forensic facial-recognition analysis performed by the AI can quickly sort personal images and locate those with a high likelihood of matching crime scene DNA evidence.

As the algorithm scans the database, it eliminates individuals that are incongruent with the on-file genetic material, based on characteristics such as gender or hair color. Researchers said that even if the perpetrator is not in the database, the AI will return individuals with similar genetics and physical features.

Yet, researchers warn that the face of a suspect will likely never be found using DNA alone, as the shape of the human face is also determined by factors such as diet, age, environment and socioeconomic background.

“The more genes we identify, the more accurate this method becomes, and it will only continue to improve as our knowledge of the relevant genes grows,” explained KU Leuven engineer and study author Peter Claes.

Such a system, according to researchers, requires large facial image databases, like ID card or license databases, to be effective. Researchers cautioned there are ethical and legal considerations to be addressed when accessing databases that hold private information.

The research is detailed in the journal Nature Communications.

To contact the author of this article, email mdonlon@globalspec.com