Forensic Experts Use Blood Flow Detection to Identify Deepfakes

By Netvora Tech News

---

The Netherlands Forensic Institute (NFI) has developed a method to identify deepfakes by using blood flow detection. The technique involves monitoring the expansion and contraction of blood vessels in the face. According to the NFI, when blood flow increases and decreases with each heartbeat, a subtle color change occurs that can be measured. This change in blood flow is not present in deepfakes, making it a potential indicator of authenticity. The NFI notes that the vessels around the eyes, forehead, and jaw are particularly suitable for measuring this expansion, as they are close to the surface of the skin. The impact of a deepfake can be significant, both for the individual and society. In times of heightened political tension, the ability to detect deepfakes is more urgent than ever. The NFI is currently validating the scientific validity of the method. The idea of using blood flow detection to measure someone's heart rate is not new. In 2009, the Massachusetts Institute of Technology (MIT) published a study that showed it was possible to measure a person's heart rate based on the blood vessels in their face. However, the image compression technology at the time was not sufficient to achieve this. "Now, with better compression techniques, we can measure even the slightest changes in blood flow," says NFI researcher Zeno Geradts. To the best of our knowledge, blood flow detection is not currently used in forensic research. "We are still working on the scientific validation, but it has the potential to be a valuable addition to existing methods," Geradts adds. The researcher will present the new method at the European Academy of Forensic Science (EAFS) 2025 conference in Dublin from May 26 to 30.