The Role of Deep Learning and Gait Analysis in Parkinson’s Disease: A Systematic Review DOI Creative Commons
Alessandra Franco, Michela Russo, Marianna Amboni

et al.

Sensors, Journal Year: 2024, Volume and Issue: 24(18), P. 5957 - 5957

Published: Sept. 13, 2024

Parkinson's disease (PD) is the second most common movement disorder in world. It characterized by motor and non-motor symptoms that have a profound impact on independence quality of life people affected disease, which increases caregivers' burdens. The use quantitative gait data with PD deep learning (DL) approaches based are emerging as increasingly promising methods to support aid clinical decision making, aim providing objective diagnosis, well an additional tool for monitoring. This will allow early detection assessment progression, implementation therapeutic interventions. In this paper, authors provide systematic review DL techniques recently proposed analysis using Preferred Reporting Items Systematic Reviews Meta-Analyses (PRISMA) guidelines. Scopus, PubMed, Web Science databases were searched across interval six years (between 2018, when first article was published, 2023). A total 25 articles included review, reports studies patients both wearable non-wearable sensors. Additionally, these employed networks classification, monitoring purposes. demonstrate there wide employment field convolutional neural analyzing signals from sensors pose estimation motion videos. addition, discuss current difficulties highlight future solutions progression.

Language: Английский

Biomechanics Parameters of Gait Analysis to Characterize Parkinson’s Disease: A Scoping Review DOI Creative Commons
Michela Russo, Marianna Amboni,

Noemi Pisani

et al.

Sensors, Journal Year: 2025, Volume and Issue: 25(2), P. 338 - 338

Published: Jan. 9, 2025

Parkinson's disease (PD) is characterized by a slow, short-stepping, shuffling gait pattern caused combination of motor control limitations due to reduction in dopaminergic neurons. Gait disorders are indicators global health, cognitive status, and risk falls increase with progression. Therefore, the use quantitative information on mechanisms PD patients promising approach, particularly for monitoring potentially informing therapeutic interventions, though it not yet well-established tool early diagnosis or direct assessment Over years, many studies have investigated spatiotemporal parameters that altered pattern, while kinematic kinetic more limited. A scoping review was performed according PRISMA guidelines. The Scopus PubMed databases were searched between 1999 2023. total 29 articles included reported changes under different conditions: single free walking, sequential task, dual task. main findings our highlighted optoelectronic systems recording force plates measuring parameters, their high accuracy. Most analyses been conducted at self-selected walking speeds capture natural movement, although also examined various conditions. results indicated experience alterations range motion hip, knee, ankle joints, as well power generated/absorbed extensor/flexor moments. These suggest may be effectively understood using parameters.

Language: Английский

Citations

1
The Role of Deep Learning and Gait Analysis in Parkinson’s Disease: A Systematic Review DOI Creative Commons
Alessandra Franco, Michela Russo, Marianna Amboni

et al.

Sensors, Journal Year: 2024, Volume and Issue: 24(18), P. 5957 - 5957

Published: Sept. 13, 2024

Parkinson's disease (PD) is the second most common movement disorder in world. It characterized by motor and non-motor symptoms that have a profound impact on independence quality of life people affected disease, which increases caregivers' burdens. The use quantitative gait data with PD deep learning (DL) approaches based are emerging as increasingly promising methods to support aid clinical decision making, aim providing objective diagnosis, well an additional tool for monitoring. This will allow early detection assessment progression, implementation therapeutic interventions. In this paper, authors provide systematic review DL techniques recently proposed analysis using Preferred Reporting Items Systematic Reviews Meta-Analyses (PRISMA) guidelines. Scopus, PubMed, Web Science databases were searched across interval six years (between 2018, when first article was published, 2023). A total 25 articles included review, reports studies patients both wearable non-wearable sensors. Additionally, these employed networks classification, monitoring purposes. demonstrate there wide employment field convolutional neural analyzing signals from sensors pose estimation motion videos. addition, discuss current difficulties highlight future solutions progression.

Language: Английский

Citations

3