Proof of Concept in Artificial-Intelligence-Based Wearable Gait Monitoring for Parkinson’s Disease Management Optimization
Parkinson’s disease (PD) is the second most common progressive neurodegenerative disorder, affecting 6.2 million patients and causing disability and decreased quality of life. The research is oriented nowadays toward artificial intelligence (AI)-based wearables for early diagnosis and long-term PD m...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-03-01
|
Series: | Biosensors |
Subjects: | |
Online Access: | https://www.mdpi.com/2079-6374/12/4/189 |
_version_ | 1797436656212508672 |
---|---|
author | Robert Radu Ileșan Claudia-Georgiana Cordoș Laura-Ioana Mihăilă Radu Fleșar Ana-Sorina Popescu Lăcrămioara Perju-Dumbravă Paul Faragó |
author_facet | Robert Radu Ileșan Claudia-Georgiana Cordoș Laura-Ioana Mihăilă Radu Fleșar Ana-Sorina Popescu Lăcrămioara Perju-Dumbravă Paul Faragó |
author_sort | Robert Radu Ileșan |
collection | DOAJ |
description | Parkinson’s disease (PD) is the second most common progressive neurodegenerative disorder, affecting 6.2 million patients and causing disability and decreased quality of life. The research is oriented nowadays toward artificial intelligence (AI)-based wearables for early diagnosis and long-term PD monitoring. Our primary objective is the monitoring and assessment of gait in PD patients. We propose a wearable physiograph for qualitative and quantitative gait assessment, which performs bilateral tracking of the foot biomechanics and unilateral tracking of arm balance. Gait patterns are assessed by means of correlation. The surface plot of a correlation coefficient matrix, generated from the recorded signals, is classified using convolutional neural networks into physiological or PD-specific gait. The novelty is given by the proposed AI-based decisional support procedure for gait assessment. A proof of concept of the proposed physiograph is validated in a clinical environment on five patients and five healthy controls, proving to be a feasible solution for ubiquitous gait monitoring and assessment in PD. PD management demonstrates the complexity of the human body. A platform empowering multidisciplinary, AI-evidence-based decision support assessments for optimal dosing between drug and non-drug therapy could lay the foundation for affordable precision medicine. |
first_indexed | 2024-03-09T11:05:46Z |
format | Article |
id | doaj.art-08a613a9d18040a0b99f008f299b687d |
institution | Directory Open Access Journal |
issn | 2079-6374 |
language | English |
last_indexed | 2024-03-09T11:05:46Z |
publishDate | 2022-03-01 |
publisher | MDPI AG |
record_format | Article |
series | Biosensors |
spelling | doaj.art-08a613a9d18040a0b99f008f299b687d2023-12-01T00:57:34ZengMDPI AGBiosensors2079-63742022-03-0112418910.3390/bios12040189Proof of Concept in Artificial-Intelligence-Based Wearable Gait Monitoring for Parkinson’s Disease Management OptimizationRobert Radu Ileșan0Claudia-Georgiana Cordoș1Laura-Ioana Mihăilă2Radu Fleșar3Ana-Sorina Popescu4Lăcrămioara Perju-Dumbravă5Paul Faragó6Department of Neurology and Pediatric Neurology, Faculty of Medicine, University of Medicine and Pharmacy “Iuliu Hatieganu” Cluj-Napoca, 400012 Cluj-Napoca, RomaniaBases of Electronics Department, Faculty of Electronics, Telecommunications and Information Technology, Technical University of Cluj-Napoca, 400114 Cluj-Napoca, RomaniaBases of Electronics Department, Faculty of Electronics, Telecommunications and Information Technology, Technical University of Cluj-Napoca, 400114 Cluj-Napoca, RomaniaComputer Science, Faculty of Mathematics and Computer Science, West University of Timișoara, 300223 Timișoara, RomaniaDepartment of Neurology and Pediatric Neurology, Faculty of Medicine, University of Medicine and Pharmacy “Iuliu Hatieganu” Cluj-Napoca, 400012 Cluj-Napoca, RomaniaDepartment of Neurology and Pediatric Neurology, Faculty of Medicine, University of Medicine and Pharmacy “Iuliu Hatieganu” Cluj-Napoca, 400012 Cluj-Napoca, RomaniaBases of Electronics Department, Faculty of Electronics, Telecommunications and Information Technology, Technical University of Cluj-Napoca, 400114 Cluj-Napoca, RomaniaParkinson’s disease (PD) is the second most common progressive neurodegenerative disorder, affecting 6.2 million patients and causing disability and decreased quality of life. The research is oriented nowadays toward artificial intelligence (AI)-based wearables for early diagnosis and long-term PD monitoring. Our primary objective is the monitoring and assessment of gait in PD patients. We propose a wearable physiograph for qualitative and quantitative gait assessment, which performs bilateral tracking of the foot biomechanics and unilateral tracking of arm balance. Gait patterns are assessed by means of correlation. The surface plot of a correlation coefficient matrix, generated from the recorded signals, is classified using convolutional neural networks into physiological or PD-specific gait. The novelty is given by the proposed AI-based decisional support procedure for gait assessment. A proof of concept of the proposed physiograph is validated in a clinical environment on five patients and five healthy controls, proving to be a feasible solution for ubiquitous gait monitoring and assessment in PD. PD management demonstrates the complexity of the human body. A platform empowering multidisciplinary, AI-evidence-based decision support assessments for optimal dosing between drug and non-drug therapy could lay the foundation for affordable precision medicine.https://www.mdpi.com/2079-6374/12/4/189artificial intelligencesensorsconvolutional neural networksParkinson’s diseasebiomedical monitoringaccelerometer |
spellingShingle | Robert Radu Ileșan Claudia-Georgiana Cordoș Laura-Ioana Mihăilă Radu Fleșar Ana-Sorina Popescu Lăcrămioara Perju-Dumbravă Paul Faragó Proof of Concept in Artificial-Intelligence-Based Wearable Gait Monitoring for Parkinson’s Disease Management Optimization Biosensors artificial intelligence sensors convolutional neural networks Parkinson’s disease biomedical monitoring accelerometer |
title | Proof of Concept in Artificial-Intelligence-Based Wearable Gait Monitoring for Parkinson’s Disease Management Optimization |
title_full | Proof of Concept in Artificial-Intelligence-Based Wearable Gait Monitoring for Parkinson’s Disease Management Optimization |
title_fullStr | Proof of Concept in Artificial-Intelligence-Based Wearable Gait Monitoring for Parkinson’s Disease Management Optimization |
title_full_unstemmed | Proof of Concept in Artificial-Intelligence-Based Wearable Gait Monitoring for Parkinson’s Disease Management Optimization |
title_short | Proof of Concept in Artificial-Intelligence-Based Wearable Gait Monitoring for Parkinson’s Disease Management Optimization |
title_sort | proof of concept in artificial intelligence based wearable gait monitoring for parkinson s disease management optimization |
topic | artificial intelligence sensors convolutional neural networks Parkinson’s disease biomedical monitoring accelerometer |
url | https://www.mdpi.com/2079-6374/12/4/189 |
work_keys_str_mv | AT robertraduilesan proofofconceptinartificialintelligencebasedwearablegaitmonitoringforparkinsonsdiseasemanagementoptimization AT claudiageorgianacordos proofofconceptinartificialintelligencebasedwearablegaitmonitoringforparkinsonsdiseasemanagementoptimization AT lauraioanamihaila proofofconceptinartificialintelligencebasedwearablegaitmonitoringforparkinsonsdiseasemanagementoptimization AT raduflesar proofofconceptinartificialintelligencebasedwearablegaitmonitoringforparkinsonsdiseasemanagementoptimization AT anasorinapopescu proofofconceptinartificialintelligencebasedwearablegaitmonitoringforparkinsonsdiseasemanagementoptimization AT lacramioaraperjudumbrava proofofconceptinartificialintelligencebasedwearablegaitmonitoringforparkinsonsdiseasemanagementoptimization AT paulfarago proofofconceptinartificialintelligencebasedwearablegaitmonitoringforparkinsonsdiseasemanagementoptimization |