Researchmetabolomicsforensic pathologymachine learning
Researchers Use AI To Estimate Time-of-Death
9.2
Researchers at Linköping University and the Swedish National Board of Forensic Medicine publish in Nature Communications that they trained an AI model on metabolite profiles to estimate post-mortem interval. The model used 4,876 autopsy blood samples drawn from a 45,000-sample biobank and achieves about one-day precision for deaths up to 13 days. The approach could improve forensic timelines and investigative prioritization.
Key Points
- 1Trained AI on 4,876 autopsy blood samples to estimate post-mortem interval from metabolite profiles.
- 2Achieved roughly one-day precision for deaths up to 13 days, outperforming traditional temperature and rigor metrics.
- 3Allows forensic labs to narrow timelines and prioritize witness or resource allocation in complex investigations.
Scoring Rationale
Strong peer-reviewed results and clear forensic applicability justify a high score, but broader external validation is needed.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems