Source study found

Story checked

AI-powered heart MRI measurements better predict patients' survival (opens in a new tab)

news-medical.net · 2026-11-02

Short answerEvidenceSource

Short answer

Mixed

Mixed.

One claim goes further than the study. 2 other points were not covered by the paper.

3 supported
1 overstated
2 not covered

Checked against the study summary. The full text wasn't available, so some details couldn't be settled either way.

Share this check

Follow the evidence trail

The story

AI-powered heart MRI measurements better predict patients' survival

news-medical.net · 2026-11-02

The story’s checkable claims.

Read the original story

NewsLink checks it

Mixed

One claim overstates the study. Three of six check out. Two claims the study doesn't address.

3 supported
1 overstated
2 not covered

Open claim evidence

The source study

Performance of artificial intelligence versus clinician left-ventricular segmentation of cardiac MRI for mortality prediction: a retrospective, multicentre, prognostic study

The Lancet. Digital Health · 2027

Open source record Open in Tessa

Then inspect each claim

Evidence layer

Claim by claim

Each claim gets a verdict. Expand it to see the evidence directly below.

Reading mode

Scan verdicts. Open evidence only when needed.

Browse by verdict

6 claims in this story

Showing all 6 claimsChoose a verdict to focus the list.

Claim 1 of 6OverstatedThe research focused on ejection fraction, described as the most important measure of heart function and a 'gatekeeper' for many treatments.View evidenceHide evidence

As statedmost important measure of heart function

Why this verdict

The paper profile supports that ejection fraction is a key measure used for clinical decision-making and risk stratification, and that EF was central to the study. However, the supplied abstract-level profile does not support the superlative “the most important measure of heart function” or the stronger “gatekeeper for many treatments” wording. Those statements go beyond the profiled evidence.

Study evidence

AI-derived ejection fraction provided marginally better prognostic performance for all-cause mortality than clinician-derived ejection fraction in a large retrospective multicentre CMR cohort.Unadjusted: AIC 57,936 vs 58,097 (ΔAIC = 161 favoring AI); C-index 0.669 vs 0.637 (ΔC = 0.033, 95% CI 0.026–0.041). Adjusted for age and sex: AIC 55,684 vs 55,742 (ΔAIC = 59); C-index 0.781 vs 0.777 (ΔC = 0.004, 95% CI 0.002–0.006).

Claim 2 of 6Not coveredThe AI tool analyzed scans in approximately 2–3 seconds per patient, compared with previous expert clinician analysis that typically takes around 13 minutes.View evidenceHide evidence

As stated2-3 seconds per patient versus around 13 minutes

Why this verdict

The abstract-level profile describes AI segmentation and clinician-derived measurements, but it does not report scan-analysis time, 2–3 seconds per patient, or a 13-minute clinician comparison. This may be present outside the abstract, but it is not verifiable from the supplied abstract-depth profile.

Study evidence

Claim 3 of 6Not coveredThe researchers stress that human oversight remains essential, noting AI errors were uncommon at around 1% at the patient level and affected 0.08% of images.View evidenceHide evidence

As statedaround 1% at the patient level; 0.08% of images

Why this verdict

The profile supports the general idea that clinician oversight is recommended, but the supplied abstract-level evidence does not report the quoted AI error rates of around 1% at the patient level or 0.08% of images. Therefore the full quantitative claim is not verifiable at this evidence depth.

Study evidence

Claim 4 of 6SupportedAI measurements of heart function from MRI scans are more strongly linked to how long patients live after their scan than measurements made by clinicians.View evidenceHide evidence

As statedmore strongly linked

Why this verdict

The abstract-level profile supports an associational, prognostic framing: AI-derived ejection fraction from cardiac MRI had better model fit and discrimination for all-cause mortality than clinician-derived ejection fraction. The effect is described in the paper profile as marginal/small, so the headline is acceptable because it says “linked” rather than causal, but the magnitude should be understood as modest.

Study evidence

Study evidence

Unadjusted model comparisons showed better overall model fit and discrimination for AI-derived ejection fraction versus clinician-derived ejection fraction (AIC 57,936 vs 58,097; ΔAIC = 161 favoring AI; C-index 0.669 vs 0.637; ΔC = 0.033, 95% CI 0.026–0.041).ΔAIC = 161; ΔC = 0.033 (95% CI 0.026–0.041)

Claim 5 of 6SupportedResearchers analyzed heart MRI scans from 27,382 patients in the UK and United States, with an average follow-up of more than five years and 3,458 deaths.View evidenceHide evidence

As stated27,382 patients; more than five years; 3,458 deaths

Why this verdict

The profile supports the cohort size of 27,382 patients, UK and USA sites, median follow-up of 5.6 years, and 3,458 deaths. The story’s “average follow-up” wording is slightly imprecise because the profile reports median follow-up, but the substantive claim of more than five years of follow-up is supported.

Study evidence

Study evidence

Claim 6 of 6SupportedAI-derived ejection fraction measurements were slightly but significantly better at predicting which patients were at increased risk of death than clinician-derived measurements, after accounting for age and sex.View evidenceHide evidence

As statedslightly but significantly better

Why this verdict

The profile directly supports this claim. After adjustment for age and sex, AI-derived EF had better model fit and marginally better discrimination than clinician-derived EF, with C-index 0.781 vs 0.777 and ΔC 0.004. The story’s “slightly but significantly better” phrasing matches the small, statistically supported difference.

Study evidence

Study evidence

Then look for missing context

Context layer

What the story left out

Important study details the story did not include.

Incremental-value analysis: adding AI EF to clinician EF improved model fit more than adding clinician EF to AI EF
The paper profile treats nested-model incremental value as a secondary contribution, including likelihood-ratio tests. The story’s presented claims do not clearly report this incremental-model finding, although they do cover the broader AI-versus-clinician comparison.
From retrospective multicentre prognostic observational study
Prospective evaluation needed before broader clinical adoption
The profile states that the authors recommend prospective studies and that clinical impact remains uncertain. The story caveats mention oversight and verification but do not appear to mention the need for prospective validation.
From Retrospective multicentre prognostic cohort; retrospective multicentre prognostic observational study

6 things the story did carry across

Retrospective multicentre observational prognostic cohort design
Large cohort and mortality follow-up: 27,382 cardiac MRI patients, median follow-up 5.6 years, 3,458 deaths
Main finding: AI-derived ejection fraction was marginally better than clinician-derived ejection fraction for all-cause mortality prediction
Age- and sex-adjusted comparison showed only a very small discrimination gain
Need for clinician oversight and contextual interpretation
Clinical impact is uncertain because improvements in discrimination were numerically small

Then read the study layer

Study layer

Study at a glance

Scan the study first. Expand only the parts you want to inspect.

Pieces of work

Evidence read

study summary

Lead result

secondary data

1Lead resultsecondary dataCompare the prognostic performance (all-cause mortality prediction) of AI-derived versus clinician-derived left-ventricular volume metrics (especially ejection fraction) from cardiac MRI in a large retrospective, multicentre cohort.Retrospective multicentre prognostic cohortExpand

In plain English

Retrospective, multicentre prognostic cohort study (27,382 consecutive CMR patients from Barts Heart Centre and UPMC) comparing AI deep-learning versus routine clinician left-ventricular segmentation–derived ejection fraction for prediction of all-cause mortality using Cox proportional hazards models; AI-derived ejection fraction produced marginally better model fit (AIC) and discrimination (C-index) than clinician-derived ejection fraction, including after adjustment for age and sex.

Key findings

AI-derived ejection fraction provided marginally better prognostic performance for all-cause mortality than clinician-derived ejection fraction in a large retrospective multicentre CMR cohort.Unadjusted: AIC 57,936 vs 58,097 (ΔAIC = 161 favoring AI); C-index 0.669 vs 0.637 (ΔC = 0.033, 95% CI 0.026–0.041). Adjusted for age and sex: AIC 55,684 vs 55,742 (ΔAIC = 59); C-index 0.781 vs 0.777 (ΔC = 0.004, 95% CI 0.002–0.006).

What this piece can’t prove

Retrospective, observational design (cohort assembled from existing reporting archives).
Reported improvements in discrimination are numerically small (described as marginal), so clinical impact is uncertain without further prospective study.

1 further detail could not be confirmed from the summary.

2secondary dataEstimate the incremental prognostic value of AI-derived ejection fraction over clinician-derived ejection fraction (and vice versa), including in models adjusted for age and sex.retrospective multicentre prognostic observational studyExpand

In plain English

The paper reports nested Cox model comparisons to quantify the incremental prognostic value of AI-derived left-ventricular ejection fraction (EF) versus clinician-derived EF for all-cause mortality. Using likelihood-ratio tests, AIC, and C-indices, the authors found that adding AI EF to clinician EF produced a substantial, statistically significant improvement in model fit, whereas adding clinician EF to AI produced only a small improvement; results were similar but attenuated after adjustment for age and sex.

Key findings

Unadjusted model comparisons showed better overall model fit and discrimination for AI-derived ejection fraction versus clinician-derived ejection fraction (AIC 57,936 vs 58,097; ΔAIC = 161 favoring AI; C-index 0.669 vs 0.637; ΔC = 0.033, 95% CI 0.026–0.041).ΔAIC = 161; ΔC = 0.033 (95% CI 0.026–0.041)
Incremental-value testing (likelihood-ratio tests) indicated that adding AI EF to a model containing clinician EF produced a large, statistically significant improvement (χ2 = 168, p < 0.0001), whereas adding clinician EF to a model containing AI EF produced only a small improvement (χ2 = 7.5, p = 0.023).LR test χ2 = 168 (p < 0.0001) for AI added to clinician; χ2 = 7.5 (p = 0.023) for clinician added to AI

What this piece can’t prove

Retrospective, observational design (authors recommend prospective studies).
Abstract does not report detailed prespecification of nested-model comparisons or additional covariate handling beyond age and sex.

1 further detail could not be confirmed from the summary.

Finally, the search trail

Method layer