Breaking News:AI can determine time of death from a single blood sample– What Just Happened

Artificial intelligence (AI) has now demonstrated that it can determine time of death from a single blood sample with day-level accuracy, even nearly two weeks after life has ended. 

That sharper clock can redirect a criminal investigation, tightening alibis and narrowing the search for witnesses when uncertainty once stretched for days.

AI reveals time of death

Blood samples drawn during an autopsy continues to carry a measurable record of how long a body has been without life. 

Working with thousands of such cases, Dr. Rasmus Magnusson at Linköping University (LiU) demonstrated that chemical changes can be translated into a reliable estimate of elapsed time. 

Across deaths spanning nearly two weeks, the signal held steady enough to distinguish one day from the next.

That reliability highlights the AI method’s potential and raises the question of how it stacks up against traditional forensic tools.

Blood chemistry shifts after death

After death, cells lose control of their internal chemistry, and metabolites – the small molecules produced during normal cellular reactions – begin to shift rapidly.

Some of these molecules break down and others build up as key proteins stop working and cell walls start leaking.

In whole blood, that shifting mix mirrors changes across multiple organs, meaning a single sample can contain a wealth of timing information.

Patterns alone cannot close a case, but they can strengthen other evidence when traditional methods for estimating time since death reach their limits.

Old methods lose accuracy

Body cooling, muscle stiffening, and chemical changes in eye fluid often guide early estimates of time since death.

Forensic pathologists call that span the post-mortem interval (PMI), the time between death and examination of the body.

Beyond the first 48 hours, even trusted lab measures like potassium in eye fluid grow less precise, widening the time window investigators must work within.

Once that early window closes, investigators are often left with wide time ranges that can blur alibis and obscure crucial movements.

Training on real cases

Sweden’s National Board of Forensic Medicine supplied thousands of autopsy cases, including 4,876 people with a recorded time since death.

During routine drug testing, analysts collected metabolomics, large-scale measurement of many metabolites at once, from blood drawn during autopsy.

Instead of relying on a single marker, the computer model learned patterns across hundreds of chemicals.

Even with messy real-world deaths, the AI approach held onto a clear signal that linked chemistry to elapsed time.

Recycling routine blood tests

Forensic labs already scan postmortem blood for drugs, and that same scan can capture natural breakdown chemicals.

High-resolution mass spectrometry, a chemical scanner that sorts molecules by weight, produced the raw patterns the model learned.

By repurposing measurements already collected in routine workups, the team sidestepped the need for extra tests in each case – an important advantage for busy morgues.

That shifts the real challenge to maintaining consistent data quality and standardized handling, rather than investing in new equipment or overhauling established protocols.

Model tested on new cases

Validation mattered because one lab’s machine and habits can differ from the next, changing what blood data looks like.

To avoid that risk, the team tested the AI model on 512 new cases measured in another year.

Different instruments still detected enough overlapping chemicals for the model to keep working without retraining.

That kind of consistency makes the method more realistic for agencies that want similar results across places.

“We knew that many external factors affect body decomposition and were surprised that the signal from the body’s metabolites was so strong when it comes to predicting the post-mortem interval,” said Dr. Elin Nyman, a systems biology researcher at LiU.

Smaller laboratories often lack huge databases, so a method that needs only a modest sample matters.

With fewer examples, the computer can still learn the same chemical pattern, though its estimates may widen.

“A few hundred individuals are enough to build corresponding models, which makes our method useful even in laboratories worldwide that don’t have access to as much data,” said Magnusson.

That portability could let forensic labs compare results across countries instead of treating each case as a one-off.

Blood test narrows death time window

Police work often turns on a narrow timeline, so a lab result that narrows the window can redirect an investigation.

Back at the crime scene, detectives can match the estimate against phone records, cameras, and the last time someone was seen alive.

“This enables us to assess the actual time of death of an individual, which is very important in forensic investigations, but also to the work of the police,” said Henrik Green, professor of forensic sciences at LiU.

Next, the team plans to work with cases that include exact times of death, allowing future models to narrow estimates to specific parts of the day.

That progression could turn routine blood chemistry into a practical timing tool that remains reliable after traditional signs have faded.

Whether it becomes a standard part of autopsy practice or stays a niche technique will depend on rigorous validation and transparent reporting.

The study is published in the journal Nature Communications.

—–

Like what you read? Subscribe to our newsletter for engaging articles, exclusive content, and the latest updates.

Check us out on EarthSnap, a free app brought to you by Eric Ralls and Earth.com.

—–