− HEALTH We are the last generation that will influence what wellbeing means before AI systems optimize that definitio…− HEALTH Abstract There is much discussion of the false outputs that generative AI systems such as ChatGPT, Claude, Ge…± SCIENCE This article presents a Delphi consensus developed by a panel of editors-in-chief of anaesthesiology and pain…+ EDUCATION The study investigated how innovative, AI-supported teaching methods relate to the mathematics problem-solvin…+ EDUCATION As generative AI becomes increasingly integrated into higher education, understanding how students engage wit…+ HEALTH In the complex and multidimensional field of medicine, multimodal data are prevalent and crucial for informed…± SCIENCE This paper explores the evolution of geoscientific inquiry, tracing the progression from traditional physics-…− POLICY This paper conducts a systematic review and interdisciplinary analysis of the ethical challenges of generativ…
TruaceTracing the truth around AIMonday, July 20, 2026
Subscribe
Health·The Trace·Model-prefilled trace·Published 2026-07-20

AI-based coronary artery calcium scoring on routine non-gated chest CT for cardiovascular risk stratification

Source article: Quantifying the impact of slice thickness on cardiovascular risk stratification in lung cancer screening: a multi-center "RESCUE" study

Background: Patients undergoing routine non-gated chest computed tomography (CT) for health checkups or atypical chest discomfort often present with a coronary artery calcium (CAC) score of zero on standard 5.0 mm reconstructions. We hypothesized that these thick slices obscure mild calcification due to partial volume effects (PVEs), which could be recovered by retrospective analysis of native thin-slice images. This study aimed to quantify the rate of unrecognized coronary calcification on standard thick-slice…

TRV-2026-0314Peer-reviewedPermanent record — cite & verify
Trace impact reading

Contested: both sides are scored from claims and sources, not community votes.

P 69The P score combines the specificity and measured human impact of the grounded problem claim with the strength of this Trace’s cited sources.G 68The G score combines the specificity and measured human impact of the grounded gain claim with the strength of this Trace’s cited sources.
Quantifying the impact of slice thickness on cardiovascular risk stratification in lung cancer screening: a multi-center "RESCUE" study

"CT scan of lung cancer with cavitation" by Jmarchn is marked with CC0 1.0. To view the terms, visit https://creativecommons.org/publicdomain/zero/1.0/deed.en/.

The quick read

By July 2026, researchers had tested whether native thin-slice images could recover coronary calcification missed on standard 5.0 mm chest CT. Using a validated deep learning algorithm to quantify CAC on paired reconstructions, they found 19.0% of internal cohort patients and 10.2% of NLST participants were reclassified from CAC =0 to CAC >0 on thinner slices.

The finding matters because a CAC zero on thick slices is often used to de-escalate prevention, yet nearly a third of symptomatic zero patients in this study had obstructive disease. Thin-slice review offers a no-extra-radiation opportunistic screening gain, but most rescued cases were mild Agatston 1-99, leaving uncertainty about downstream management and outcomes.

Main points
  • Internal cohort RESCUE rate was 19.0% when comparing 1.0 mm to 5.0 mm reconstructions.
  • NLST cohort RESCUE rate was 10.2% using 2.0 mm scans.
  • 91-99% of reclassified patients fell into mild risk category Agatston 1-99.
  • 31% of symptomatic patients with CAC =0 on standard scans had obstructive CAD >50% stenosis.
  • AI correlated strongly with expert annotations r=0.956 and risk categorization showed weighted kappa 0.705-0.816.
Gain

Retrospective AI quantification of routinely available thin-slice chest CT reclassifies patients from CAC zero to positive, improving sensitivity for early subclinical atherosclerosis without additional radiation.

Problem

Standard 5.0 mm chest CT reconstructions obscure mild calcification due to partial volume effects, causing significant false-negative CAC zero assessments in routine screening.

The rundown

The multi-center RESCUE study compared paired thin- and thick-slice reconstructions from routine non-gated chest CT, using a validated deep learning algorithm and TotalSegmentator n=645 for robustness.

In symptomatic patients with CAC =0 on 5.0 mm scans, 31% had obstructive CAD >50% stenosis, and many were rescued to positive CAC status by thin-slice analysis, while overall risk categorization maintained strong agreement.

Sources

Reader signal

How should this claim be treated?

The debate