WiFreeze: multiresolution scalograms for freezing of gait detection in Parkinson’s leveraging 5G spectrum with deep learning

Ahsen Tahir, Jawad Ahmad, Syed Aziz Shah, Gordon Morison, Dawn A. Skelton, Hadi Larijani, Qammar H. Abbasi, Muhammad Ali Imran, Ryan M. Gibson

Research output: Contribution to journalArticle

13 Downloads (Pure)

Abstract

Abstract: Freezing of Gait (FOG) is an episodic absence of forward movement in Parkinson’s Disease (PD) patients and represents an onset of disabilities. FOG hinders daily activities and increases fall risk. There is high demand for automating the process of FOG detection due to its impact on health and well being of individuals. This work presents WiFreeze, a non-invasive, line of sight and lighting agnostic WiFi based sensing system, which exploits ambient 5G spectrum for detection and classification of FOG. The core idea is to utilize the amplitude variations of wireless Channel State Information (CSI) to differentiate between FOG and activities of daily life. A total of 225 events with 45 FOG cases are captured from 15 patients with the help of 30 subcarriers and classification is performed with a deep neural network. Multiresolution scalograms are proposed for time-frequency signatures of human activities due to their ability to capture and detect transients in CSI signals caused by transitions in human movements. A very deep Convolutional Neural Network (CNN) VGG-8K with 8K neurons each, in fully connected layers is engineered and proposed for transfer learning with multiresolution scalogram features for detection of FOG. The proposed WiFreeze system outperforms all existing wearable and vision based systems as well as deep CNN architectures with the highest accuracy of 99.7% for FOG detection. Furthermore, the proposed system provides the highest classification accuracies of 94.3% for voluntary stop and 97.6% for walking slow activities, with an improvement of 9% and 23% respectively, over the best performing state-of-the-art deep CNN architecture.
Original languageEnglish
Article number1433
Number of pages18
JournalElectronics
Volume8
Issue number12
DOIs
Publication statusPublished - 1 Dec 2019

Fingerprint

Freezing
Channel state information
Neural networks
Network architecture
Deep learning
Neurons
Lighting
Health

Keywords

  • freezing of gate
  • deep learning
  • classification
  • WiFi sensing

Cite this

@article{2829dc52f26d4c4f98d374b26710aeb2,
title = "WiFreeze: multiresolution scalograms for freezing of gait detection in Parkinson’s leveraging 5G spectrum with deep learning",
abstract = "Abstract: Freezing of Gait (FOG) is an episodic absence of forward movement in Parkinson’s Disease (PD) patients and represents an onset of disabilities. FOG hinders daily activities and increases fall risk. There is high demand for automating the process of FOG detection due to its impact on health and well being of individuals. This work presents WiFreeze, a non-invasive, line of sight and lighting agnostic WiFi based sensing system, which exploits ambient 5G spectrum for detection and classification of FOG. The core idea is to utilize the amplitude variations of wireless Channel State Information (CSI) to differentiate between FOG and activities of daily life. A total of 225 events with 45 FOG cases are captured from 15 patients with the help of 30 subcarriers and classification is performed with a deep neural network. Multiresolution scalograms are proposed for time-frequency signatures of human activities due to their ability to capture and detect transients in CSI signals caused by transitions in human movements. A very deep Convolutional Neural Network (CNN) VGG-8K with 8K neurons each, in fully connected layers is engineered and proposed for transfer learning with multiresolution scalogram features for detection of FOG. The proposed WiFreeze system outperforms all existing wearable and vision based systems as well as deep CNN architectures with the highest accuracy of 99.7{\%} for FOG detection. Furthermore, the proposed system provides the highest classification accuracies of 94.3{\%} for voluntary stop and 97.6{\%} for walking slow activities, with an improvement of 9{\%} and 23{\%} respectively, over the best performing state-of-the-art deep CNN architecture.",
keywords = "freezing of gate, deep learning, classification, WiFi sensing",
author = "Ahsen Tahir and Jawad Ahmad and {Aziz Shah}, Syed and Gordon Morison and Skelton, {Dawn A.} and Hadi Larijani and Abbasi, {Qammar H.} and Imran, {Muhammad Ali} and Gibson, {Ryan M.}",
note = "Acceptance in SAN/ from webpage OA article",
year = "2019",
month = "12",
day = "1",
doi = "10.3390/electronics8121433",
language = "English",
volume = "8",
number = "12",

}

WiFreeze: multiresolution scalograms for freezing of gait detection in Parkinson’s leveraging 5G spectrum with deep learning. / Tahir, Ahsen; Ahmad, Jawad; Aziz Shah, Syed; Morison, Gordon; Skelton, Dawn A.; Larijani, Hadi; Abbasi, Qammar H.; Imran, Muhammad Ali; Gibson, Ryan M.

In: Electronics, Vol. 8, No. 12, 1433, 01.12.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - WiFreeze: multiresolution scalograms for freezing of gait detection in Parkinson’s leveraging 5G spectrum with deep learning

AU - Tahir, Ahsen

AU - Ahmad, Jawad

AU - Aziz Shah, Syed

AU - Morison, Gordon

AU - Skelton, Dawn A.

AU - Larijani, Hadi

AU - Abbasi, Qammar H.

AU - Imran, Muhammad Ali

AU - Gibson, Ryan M.

N1 - Acceptance in SAN/ from webpage OA article

PY - 2019/12/1

Y1 - 2019/12/1

N2 - Abstract: Freezing of Gait (FOG) is an episodic absence of forward movement in Parkinson’s Disease (PD) patients and represents an onset of disabilities. FOG hinders daily activities and increases fall risk. There is high demand for automating the process of FOG detection due to its impact on health and well being of individuals. This work presents WiFreeze, a non-invasive, line of sight and lighting agnostic WiFi based sensing system, which exploits ambient 5G spectrum for detection and classification of FOG. The core idea is to utilize the amplitude variations of wireless Channel State Information (CSI) to differentiate between FOG and activities of daily life. A total of 225 events with 45 FOG cases are captured from 15 patients with the help of 30 subcarriers and classification is performed with a deep neural network. Multiresolution scalograms are proposed for time-frequency signatures of human activities due to their ability to capture and detect transients in CSI signals caused by transitions in human movements. A very deep Convolutional Neural Network (CNN) VGG-8K with 8K neurons each, in fully connected layers is engineered and proposed for transfer learning with multiresolution scalogram features for detection of FOG. The proposed WiFreeze system outperforms all existing wearable and vision based systems as well as deep CNN architectures with the highest accuracy of 99.7% for FOG detection. Furthermore, the proposed system provides the highest classification accuracies of 94.3% for voluntary stop and 97.6% for walking slow activities, with an improvement of 9% and 23% respectively, over the best performing state-of-the-art deep CNN architecture.

AB - Abstract: Freezing of Gait (FOG) is an episodic absence of forward movement in Parkinson’s Disease (PD) patients and represents an onset of disabilities. FOG hinders daily activities and increases fall risk. There is high demand for automating the process of FOG detection due to its impact on health and well being of individuals. This work presents WiFreeze, a non-invasive, line of sight and lighting agnostic WiFi based sensing system, which exploits ambient 5G spectrum for detection and classification of FOG. The core idea is to utilize the amplitude variations of wireless Channel State Information (CSI) to differentiate between FOG and activities of daily life. A total of 225 events with 45 FOG cases are captured from 15 patients with the help of 30 subcarriers and classification is performed with a deep neural network. Multiresolution scalograms are proposed for time-frequency signatures of human activities due to their ability to capture and detect transients in CSI signals caused by transitions in human movements. A very deep Convolutional Neural Network (CNN) VGG-8K with 8K neurons each, in fully connected layers is engineered and proposed for transfer learning with multiresolution scalogram features for detection of FOG. The proposed WiFreeze system outperforms all existing wearable and vision based systems as well as deep CNN architectures with the highest accuracy of 99.7% for FOG detection. Furthermore, the proposed system provides the highest classification accuracies of 94.3% for voluntary stop and 97.6% for walking slow activities, with an improvement of 9% and 23% respectively, over the best performing state-of-the-art deep CNN architecture.

KW - freezing of gate

KW - deep learning

KW - classification

KW - WiFi sensing

U2 - 10.3390/electronics8121433

DO - 10.3390/electronics8121433

M3 - Article

VL - 8

IS - 12

M1 - 1433

ER -