Radiomics-Derived Ischemia Burden Score (R-IBS) Calculator

Clinical calculation support · Verify results before clinical use

Input parameters

Enter the vessel-level radiomics probability and, when available, CT-FFR for each major coronary artery.

3Vessels
Dominance changes the assumed territory weights used by this calculator.
Vessel-level inputs Use the model-produced Rad-signature probability for each vessel (0–1). Do not enter stenosis percentage here.

LAD

LCx

RCA

Leave CT-FFR blank when unavailable. The source algorithm uses 1 − CT-FFR and takes the larger of that value and the radiomics probability. This combination rule is not independently validated.

Method and clinical context

Published work has evaluated a machine-learning radiomics signature derived from pericoronary tissue on coronary CT angiography to discriminate vessel-level functional ischemia, commonly defined in the cited study using CT-FFR ≤0.80. That published model is distinct from this calculator's weighted global index.

Calculation used here

For each vessel, the calculator uses the radiomics probability alone when CT-FFR is blank. When CT-FFR is supplied, it uses the larger of the radiomics probability and (1 − CT-FFR). The three effective probabilities are then multiplied by assumed dominance-based weights and summed, then multiplied by 100.

DominanceLADLCxRCA
Right-dominant50%20%30%
Left-dominant50%35%15%
Co-dominant50%25%25%

What this result does not mean

  • The output is an index from 0 to 100, not a measured or validated percentage of LV myocardium at risk.
  • The dominance weights are simplifying assumptions in the supplied source code, not weights validated by the cited radiomics study.
  • The 5% and 10% global categories are extrapolated and have not been validated for this formula.
  • The source-study vessel threshold of 0.202 should only be interpreted in the context of that model and its validation population; it is not a universal diagnostic cutoff.

References

Feng Y, Xu Z, Zhang L, et al. Machine-learning-derived radiomics signature of pericoronary tissue in coronary CT angiography associates with functional ischemia. Front Physiol. 2022;13:980996. DOI: 10.3389/fphys.2022.980996. PMID: 36225300.
Lin A, van Diemen PA, Motwani M, et al. Machine Learning From Quantitative Coronary Computed Tomography Angiography Predicts Fractional Flow Reserve-Defined Ischemia and Impaired Myocardial Blood Flow. Circ Cardiovasc Imaging. 2022;15(10):e014369. DOI: 10.1161/CIRCIMAGING.122.014369. PMID: 36252116.
Clinical disclaimer: This calculator is an educational, exploratory aid and is not a clinically validated decision tool. The global weighted formula, dominance weights, combination rule, and global index thresholds require independent validation before clinical use. Interpret CCTA, CT-FFR, and any radiomics output alongside the full clinical picture and validated diagnostic pathways.