Machine Learning-Based Peripheral Artery Disease Identification Using Laboratory-Based Gait Data

Abstract

Peripheral artery disease (PAD) manifests from atherosclerosis, which limits blood flow to the legs and causes changes in muscle structure and function, and in gait performance. PAD is underdiagnosed, which delays treatment and worsens clinical outcomes. To overcome this challenge, the purpose of this study is to develop machine learning (ML) models that distinguish individuals with and without PAD. This is the first step to using ML to identify those with PAD risk early. We built ML models based on previously acquired overground walking biomechanics data from patients with PAD and healthy controls. Gait signatures were characterized using ankle, knee, and hip joint angles, torques, and powers, as well as ground reaction forces (GRF). ML was able to classify those with and without PAD using Neural Networks or Random Forest algorithms with 89% accuracy (0.64 Matthew’s Correlation Coefficient) using all laboratory-based gait variables. Moreover, models using only GRF variables provided up to 87% accuracy (0.64 Matthew’s Correlation Coefficient). These results indicate that ML models can classify those with and without PAD using gait signatures with acceptable performance. Results also show that an ML gait signature model that uses GRF features delivers the most informative data for PAD classification.

A machine learning algorithm for peripheral artery disease prognosis using biomarker data

“…Al-Ramini and colleagues (2022) used gait measurements obtained by a camera system to diagnose PAD and obtained a similar accuracy of 87%. 26 Some of these gait measurements could also be used to quantify treatment effectiveness. 27 More recently, others have shown that a wearable accelerometer to estimate gait can provide similarly accurate measurements.…”

A biomechanical perspective on walking in patients with peripheral artery disease

“…A recent study successfully implemented a machine learning model on gait biomechanics data to identify presence of PAD. 59 Similarly, gait analysis can provide vital information to guide the selection of treatments and develop new interventions for improving the quality of life in patients with PAD.…”

Diagnosis of Disease Affecting Gait with a Body Acceleration-Based Model Using Reflected Marker Data for Training and a Wearable Accelerometer for Implementation.

“…Our findings represent considerable progress in developing a model for PAD detection using gait features that can be captured in real-world settings. Previous work from our group classified individuals as having or not having PAD using biomechanics laboratory gait measurements and applying standard ML classifiers [5]. The datasets included detailed measurements of joint angles, torques, powers, and ground reaction forces.…”

“…Moreover, the ankle-brachial index (ABI), the standard first test for PAD diagnosis, is a specialized test that is expensive, time-consuming, and only available in appropriately equipped and staffed vascular laboratories [4], [5].…”