Machine Learning-Based Peripheral Artery Disease Identification Using Laboratory-Based Gait Data
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 th...
Main Authors: | , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-09-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/22/19/7432 |
_version_ | 1797476871129006080 |
---|---|
author | Ali Al-Ramini Mahdi Hassan Farahnaz Fallahtafti Mohammad Ali Takallou Hafizur Rahman Basheer Qolomany Iraklis I. Pipinos Fadi Alsaleem Sara A. Myers |
author_facet | Ali Al-Ramini Mahdi Hassan Farahnaz Fallahtafti Mohammad Ali Takallou Hafizur Rahman Basheer Qolomany Iraklis I. Pipinos Fadi Alsaleem Sara A. Myers |
author_sort | Ali Al-Ramini |
collection | DOAJ |
description | 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. |
first_indexed | 2024-03-09T21:09:54Z |
format | Article |
id | doaj.art-2fd2484e445b4e0dbbc8a262ce91b35b |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-03-09T21:09:54Z |
publishDate | 2022-09-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-2fd2484e445b4e0dbbc8a262ce91b35b2023-11-23T21:49:08ZengMDPI AGSensors1424-82202022-09-012219743210.3390/s22197432Machine Learning-Based Peripheral Artery Disease Identification Using Laboratory-Based Gait DataAli Al-Ramini0Mahdi Hassan1Farahnaz Fallahtafti2Mohammad Ali Takallou3Hafizur Rahman4Basheer Qolomany5Iraklis I. Pipinos6Fadi Alsaleem7Sara A. Myers8Mechanical Engineering Department, University of Nebraska-Lincoln, Lincoln, NE 68588, USADepartment of Biomechanics, University of Nebraska at Omaha, Omaha, NE 6160, USADepartment of Biomechanics, University of Nebraska at Omaha, Omaha, NE 6160, USADurham School of Architectural Engineering and Construction, University of Nebraska–Lincoln, Omaha, NE 68182, USADepartment of Biomechanics, University of Nebraska at Omaha, Omaha, NE 6160, USACyber Systems Department, University of Nebraska at Kearney, Kearney, NE 68849, USADepartment of Surgery and VA Research Service, VA Nebraska-Western Iowa Health Care System, Omaha, NE 68105, USADurham School of Architectural Engineering and Construction, University of Nebraska–Lincoln, Omaha, NE 68182, USADepartment of Biomechanics, University of Nebraska at Omaha, Omaha, NE 6160, USAPeripheral 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.https://www.mdpi.com/1424-8220/22/19/7432peripheral artery diseasevascular diseasemachine learninggait analysisdeep learning |
spellingShingle | Ali Al-Ramini Mahdi Hassan Farahnaz Fallahtafti Mohammad Ali Takallou Hafizur Rahman Basheer Qolomany Iraklis I. Pipinos Fadi Alsaleem Sara A. Myers Machine Learning-Based Peripheral Artery Disease Identification Using Laboratory-Based Gait Data Sensors peripheral artery disease vascular disease machine learning gait analysis deep learning |
title | Machine Learning-Based Peripheral Artery Disease Identification Using Laboratory-Based Gait Data |
title_full | Machine Learning-Based Peripheral Artery Disease Identification Using Laboratory-Based Gait Data |
title_fullStr | Machine Learning-Based Peripheral Artery Disease Identification Using Laboratory-Based Gait Data |
title_full_unstemmed | Machine Learning-Based Peripheral Artery Disease Identification Using Laboratory-Based Gait Data |
title_short | Machine Learning-Based Peripheral Artery Disease Identification Using Laboratory-Based Gait Data |
title_sort | machine learning based peripheral artery disease identification using laboratory based gait data |
topic | peripheral artery disease vascular disease machine learning gait analysis deep learning |
url | https://www.mdpi.com/1424-8220/22/19/7432 |
work_keys_str_mv | AT alialramini machinelearningbasedperipheralarterydiseaseidentificationusinglaboratorybasedgaitdata AT mahdihassan machinelearningbasedperipheralarterydiseaseidentificationusinglaboratorybasedgaitdata AT farahnazfallahtafti machinelearningbasedperipheralarterydiseaseidentificationusinglaboratorybasedgaitdata AT mohammadalitakallou machinelearningbasedperipheralarterydiseaseidentificationusinglaboratorybasedgaitdata AT hafizurrahman machinelearningbasedperipheralarterydiseaseidentificationusinglaboratorybasedgaitdata AT basheerqolomany machinelearningbasedperipheralarterydiseaseidentificationusinglaboratorybasedgaitdata AT iraklisipipinos machinelearningbasedperipheralarterydiseaseidentificationusinglaboratorybasedgaitdata AT fadialsaleem machinelearningbasedperipheralarterydiseaseidentificationusinglaboratorybasedgaitdata AT saraamyers machinelearningbasedperipheralarterydiseaseidentificationusinglaboratorybasedgaitdata |