Plate Nº 33 · recorded November 17, 2026

Health & Medicine ResearchReported finding

AI heart-age gap linked to 46% higher relative death rate

In a 568-person Mayo Clinic study, a five-year gap in AI-estimated heart age was linked to a 46% higher relative death rate. The conference result remains preliminary.

By Elena Vasquez5 min read907 words

In brief

  1. Mayo Clinic researchers analyzed 568 adults aged at least 50 whose heart ultrasounds were recorded in 2002 to 2004.
  2. A five-year increase in the AI-estimated heart-age gap was linked to a 46% higher relative death rate after accounting for age, sex, body size and pumping ability.
  3. Follow-up records covered about two decades and combined deaths from heart and blood vessel conditions with dementia-related deaths.
  4. The team presented the preliminary results at the European Society of Cardiology's Digital & AI Summit 2026 on November 13.
  5. The analysis did not test whether treatment based on the AI estimate could prevent illness.
Could your heart be older than you? AI study finds a clue
Plate Nº 33Could your heart be older than you? AI study finds a clue — hermesmanuel / pixabay

By Elena Vasquez

A five-year increase in the gap between a person's actual age and an artificial intelligence (AI)-estimated heart age was linked to a 46% higher relative rate of death in a Mayo Clinic study of 568 adults.

The result suggests that a heart ultrasound may hold warning information beyond what doctors can see by eye. It does not show that an older-looking heart causes death, and the study did not test whether using the estimate improves care.

What did the study measure?

The Mayo Clinic team presented the work on November 13, 2026, at the European Society of Cardiology's Digital & AI Summit 2026. Mayo Clinic described the work as “preliminary conference research” rather than a completed peer-reviewed journal report, meaning a full paper checked by other scientists.

Researchers began with a heart ultrasound, a painless scan that uses sound waves to show the heart moving and help doctors judge how well it pumps blood. Rather than ask the AI to diagnose a particular disease, they had it estimate the heart's age from ultrasound recordings.

They then compared that estimate with each person's actual age. The central idea is simple: someone aged 65 can have a heart that appears much older, with health risks that may not become clear for many years.

The analysis included 568 people aged at least 50 who underwent scans in 2002 to 2004. Follow-up records covered about two decades, allowing the team to examine links between the age gap and later deaths.

The researchers counted deaths from heart and blood vessel conditions and dementia together as one group of outcomes. In this analysis, a larger gap meant the AI judged the heart to be older than its owner.

Takeaway: The AI estimated heart age; it did not make a disease diagnosis.

What did the study find?

The greater the estimated age gap, the higher the rate of death from the combined causes during follow-up. The pattern remained after researchers accounted for participants' actual age, sex, body size and the heart's pumping ability.

With those factors taken into account, a five-year increase in the estimated heart-age gap was linked to a 46% higher relative rate of death. The researchers cautioned that this figure “does not mean a person's chance of dying automatically rises by 46 percentage points.”

The 46% figure is relative, not an absolute change in an individual's risk. It compares death rates between groups with different age gaps; it does not state how likely any one person is to die. The five-year figure also describes the size of the age difference, not the follow-up period, which lasted about two decades.

Takeaway: A larger age gap marked higher risk, but the study neither explained the cause nor calculated an individual death probability.

Why might a heart look older?

Ultrasound images contain information about the heart's shape and movement. AI systems can detect complicated patterns in those images, including patterns that may be hard for the human eye to recognize.

This study did not establish which changes in the heart produced the age estimate. It also cannot separate a shared risk from an existing illness. Blood vessel problems can affect both the heart and brain, but researchers said the signal could reflect shared risks, disease already present or something else.

The method might eventually help doctors identify people who need closer attention. Its value would come from finding useful patterns in standard imaging, not from replacing a diagnosis or standard assessment.

Takeaway: The scan may reveal hidden information, but the biology behind the signal remains uncertain.

What did the study not show?

The analysis looked backward at old scans and involved a modest sample. It counted deaths from heart and blood vessel conditions and dementia together, so readers cannot tell whether one cause drove the association more than the other.

Several questions remain before the method could become a routine screening tool:

  • Researchers did not test whether doctors could prevent illness by treating someone with a high AI heart age.
  • The 568-person sample came from people aged 50 or older, and the results need confirmation in larger and different populations.
  • Researchers must determine whether the estimate adds useful information beyond established risk factors.
  • The work has not yet appeared as a completed peer-reviewed journal report.

A larger study would need to test whether flagging an older estimated heart age leads to better decisions and, ultimately, longer, healthier lives.

Takeaway: The finding identifies a signal that needs further testing, not a ready-made screening rule.

What could happen next?

If further studies confirm the association, doctors might use AI-estimated heart age alongside standard assessments to decide who needs closer follow-up. The evidence does not yet support using the estimate by itself to guide treatment or replace conventional heart-health checks.

Related research has also explored whether AI blood tests could predict heart-disease risk up to 15 years in advance. That is separate work from the ultrasound analysis presented at the cardiology summit.

The broader lesson is that an ultrasound may contain useful information beyond what the eye can easily see. Whether AI can turn that information into better care and longer, healthier lives remains an open question.

Takeaway: Ultrasound-based heart age may become one more risk indicator, but only after larger studies and clinical testing show that it helps.

via knowridge.com (Source)

Filed under

  • artificial-intelligence-in-cardiology
  • cardiac-ultrasound
  • heart-age-estimation
  • mortality-risk
  • mayo-clinic-study
Share this article:

More from Elena Vasquez

Elena Vasquez

Show full bio

Correspondent covering money, funding and market shifts around general science news at SciBeat.

458 articles

Nearby plates

« Previous article