HomeRadiology & Imaging–Omics Scores › Vascular Remodeling Radiomic Risk Model (FRP) Calculator

Vascular Remodeling Radiomic Risk Model (FRP) Calculator

  • Perivascular Fat Radiomic Profile (FRP) Enter the FRP probability (0.00–1.00) computed by your validated imaging pipeline.
  • High-Risk Plaque (HRP) Features on CCTA HRP typically includes any of: napkin-ring sign, positive remodeling, low-attenuation plaque, or spotty calcification per society consensus.
  • Optional: Perivascular Fat Attenuation Index (FAI), HU Optional context; not used in the risk multipliers below.
  • Optional: Agatston Coronary Calcium Score (CAC) Optional context; not used in the risk multipliers below.
  • Vascular Remodeling Radiomic Risk Model: Explanation and Clinical Context
    This calculator operationalizes a published perivascular fat radiomic profile (FRP) derived from coronary CT angiography to reflect adverse perivascular adipose tissue remodeling (fibrosis and microvascular changes) that accompanies chronic coronary inflammation. The original model was trained via machine learning (random forest) using hundreds of PVAT radiomic features and validated against both biology (radiotranscriptomics) and outcomes. In cohort analyses, FRP significantly improved prediction of major adverse cardiac events beyond traditional risk factors, stenosis, calcium score and high-risk plaque characteristics. A cut-off of 0.63 defined FRP-positive status; individuals with FRP≥0.63 had ~10.8× adjusted hazard of MACE versus FRP<0.63, and those with both FRP≥0.63 and high-risk plaque features exhibited ~43× adjusted hazard versus FRP−/HRP−. For a continuous read-out, each 0.01 increase in FRP corresponded to ~12% higher adjusted hazard in the SCOT-HEART cohort. This page does not recreate the proprietary training pipeline; instead it accepts an FRP already computed by your validated workflow and translates it to risk categories/multipliers as reported. Use as a research/decision-support adjunct; clinical actions should integrate the full patient context, guideline-directed therapy, and local validation.

    References:
    Oikonomou EK, Williams MC, Kotanidis CP, et al. A novel machine learning-derived radiotranscriptomic signature of perivascular fat improves cardiac risk prediction using coronary CT angiography. Eur Heart J. 2019;40(43):3529-3543. doi:10.1093/eurheartj/ehz592.
    Shang J, et al. Predicting acute coronary syndrome within 3 years using PCAT radiomics. European Radiology. 2022. (Radiomics score predicting future ACS).
    Ayx I, et al. Radiomics in Cardiac Computed Tomography. Diagnostics. 2023;13(2):307. (Overview of cardiac CT radiomics and PVAT).
    West HW, et al. Advances in clinical imaging of vascular inflammation. Current Problems in Cardiology. 2024. (Context on PVAT/vascular inflammation imaging).

Discussion


No discussions yet. Be the first to comment.

Create Note

Notes are stored privately on your device only. No login required. Nothing is uploaded or shared.

My Notes

Report this Tool