Skip to content
DraconiticResearch
Search Sources API
203,234 studies31 sourcesUpdated 56 min ago
Peer-reviewedCC BYJournal of the Royal Society, Interface

A data-driven framework to identify restenosis-prone regions in femoral arteries from geometric and inflow waveform parameters.

  • Chotirawee Chatpattanasiri1,2
  • Federica Ninno2,3
  • Vanessa Diaz-Zuccarini1,2
  • Stavroula Balabani1,2
  1. 1Department of Mechanical Engineering, University College London , London, UK.
  2. 2UCL Hawkes Institute, University College London , London, UK.
  3. 3Department of Medical Physics and Biomedical Engineering, University College London , London, UK.
Read full textPDFDOICite

Abstract

Haemodynamic indices derived from computational fluid dynamics (CFD), such as time-averaged wall shear stress (TAWSS) and oscillatory shear index (OSI), are closely associated with restenosis risk in peripheral arterial disease (PAD). We present a machine learning-reduced order modelling (ML-ROM) framework to predict critical, restenosis-prone, haemodynamic regions accounting for both vessel geometries and inlet flow waveforms. We generated 500 parameterized virtual femoral-artery geometries, each paired with a synthetic physiologically realistic inflow waveform. CFD was used to obtain the wall shear stress (WSS) distribution, from which TAWSS and OSI were computed. Critical regions were then defined by applying different thresholding criteria based on TAWSS and OSI, as found in the literature. These were then used to construct ROMs using Proper Orthogonal Decomposition (POD), and to train ML models from which the critical regions could be reconstructed. Four ML architectures were explored: a geometric-parameter-only model and three alternatives that also incorporate inflow waveform information via Fourier coefficients, long short-term memory (LSTM), and convolutional neural networks (CNN). The Fourier models achieved the highest predictive performance, with the median values of balanced accuracy (BA) exceeding 0.92 across all critical-region definitions. The model also offered a substantial speed-up ratio of the order of 109 compared with traditional CFD.

Related

  1. Geometric Thermodynamics of Scallop Motion with Two Control Parameters
    Hisao Hayakawa
    arXivPreprintCC0Full textCondensed matterStatistical & nonlinear physicsarXiv
  2. Framework to identify defect-insensitive electron transport in topological semimetals
    Lily Jade Joyce, Michelle M. Kelley, Timothy Philicelli +2
    arXivPreprintarXiv licenceFull textCondensed matterarXiv
  3. Geometric mechanics at eleven parameters: the sixteen as a top, a fluid and a plasma
    Patrick B. Reynaud
    HALPreprintCC BY 4.0Full textGeneral physicsHAL
  4. Reconstruction of Beam Transverse Parameters in the Fermilab Side-Coupled Linac Using a Normalized Coordinate Framework
    E. V. Chen, J-P. Carneiro, R. V. Sharankova +1
    arXivPreprintCC BY 4.0Full textApplied physics & instrumentationarXiv
  5. Fully differentiable framework for inverse identification of geometry and material parameters with application to determining stress-free configuration of soft tissues
    Hyunoh Bae, Tianyi Hu, Adrian Buganza Tepole +1
    arXivPreprintarXiv licenceFull textComputational physicsarXiv

Cite

BibTeXRIS
BibTeX
@article{chatpattanasiri2026datadriven,
  title = {{A data-driven framework to identify restenosis-prone regions in femoral arteries from geometric and inflow waveform parameters.}},
  author = {Chotirawee Chatpattanasiri and Federica Ninno and Vanessa Diaz-Zuccarini and Stavroula Balabani},
  journal = {Journal of the Royal Society, Interface},
  year = {2026},
  month = oct,
  doi = {10.1098/rsif.2026.0109},
  url = {https://europepmc.org/article/MED/42869465},
  note = {CC BY},
}
RIS
TY  - JOUR
TI  - A data-driven framework to identify restenosis-prone regions in femoral arteries from geometric and inflow waveform parameters.
AU  - Chotirawee Chatpattanasiri
AU  - Federica Ninno
AU  - Vanessa Diaz-Zuccarini
AU  - Stavroula Balabani
PY  - 2026
DA  - 2026/10/01
JO  - Journal of the Royal Society, Interface
DO  - 10.1098/rsif.2026.0109
UR  - https://europepmc.org/article/MED/42869465
AB  - Haemodynamic indices derived from computational fluid dynamics (CFD), such as time-averaged wall shear stress (TAWSS) and oscillatory shear index (OSI), are closely associated with restenosis risk in peripheral arterial disease (PAD). We present a machine learning-reduced order modelling (ML-ROM) framework to predict critical, restenosis-prone, haemodynamic regions accounting for both vessel geometries and inlet flow waveforms. We generated 500 parameterized virtual femoral-artery geometries, each paired with a synthetic physiologically realistic inflow waveform. CFD was used to obtain the wall shear stress (WSS) distribution, from which TAWSS and OSI were computed. Critical regions were then defined by applying different thresholding criteria based on TAWSS and OSI, as found in the literature. These were then used to construct ROMs using Proper Orthogonal Decomposition (POD), and to train ML models from which the critical regions could be reconstructed. Four ML architectures were explored: a geometric-parameter-only model and three alternatives that also incorporate inflow waveform information via Fourier coefficients, long short-term memory (LSTM), and convolutional neural networks (CNN). The Fourier models achieved the highest predictive performance, with the median values of balanced accuracy (BA) exceeding 0.92 across all critical-region definitions. The model also offered a substantial speed-up ratio of the order of 109 compared with traditional CFD.
SN  - 1742-5689
SN  - 1742-5662
ER  -