Researchers at Linköpings universitet and the Rättsmedicinalverket have developed an artificial intelligence (AI) model that can predict the time of death with unprecedented precision, using metabolites in blood samples from thousands of real cases.The AI analyzes predictable changes in small molecules in the blood that occur after death, outperforming traditional methods such as body temperature, rigor mortis, and potassium levels in the eye. The model has been trained on nearly 4,900 samples with known postmortem intervals and can estimate the time since death with about one-day accuracy, even up to 13 days after death.This breakthrough has important implications for forensic investigations, including homicide cases, by helping authorities focus on relevant witnesses and reconstruct timelines more accurately. Future work aims to refine the model to also estimate the time of day of death.
Source:Linköpings universitet
About the Author
