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

TL;DR

This study presents a machine learning reduced order model framework that efficiently predicts restenosis-prone regions in femoral arteries using geometric and flow data, significantly speeding up traditional CFD analysis.

Contribution

The paper introduces a novel ML-ROM approach combining POD and multiple ML architectures to accurately and rapidly identify high-risk arterial regions.

Findings

Four critical-region definitions tested with high accuracy.

Four ML architectures evaluated, Fourier-based model performed best.

Speed-up of about nine orders of magnitude over CFD.

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). However, translating these insights into clinical practice may require computationally efficient approaches such as Reduced Order Model (ROM) or Machine Learning (ML). In this work, we developed an ML-ROM framework to predict critical, restenosis-prone, haemodynamic regions accounting for both vessel geometries and inlet flow waveforms. We generated 500 synthetic femoral-artery geometries parameterised by six geometric parameters, and created physiologically realistic inflow waveforms via Principal Component Analysis (PCA) of patient data. 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 threshold-based criteria to the TAWSS and OSI. Four critical-region definitions were considered: two with vessel-specific relative thresholds (TAWSS< 33rd percentile and OSI> 66nd percentile) and two with absolute thresholds (TAWSS< 0.5 Pa and OSI> 0.2). Proper orthogonal decomposition (POD) was then applied to these high-dimensional critical-region data to obtain ROMs; These were then used to train ML models from which the critical region regions could be reconstructed. Three ML architectures were explored: a Fourier-based architecture, a Long Short-term Memory (LSTM) architecture, and a Convolutional Neural Network (CNN) architecture. The Fourier models achieved the highest performance, with the median values of Balanced Accuracy (BA) exceeding 0.92 across all critical-region definitions. The ML-ROM framework also offered a substantial speed-up ratio, about nine orders of magnitude faster than traditional CFD.