Academia.eduAcademia.edu

Outline

Title
Abstract
FAQs
Figures
Introduction
Materials and Methods
Materials
Gait Motion Data Acquisition System
Data Collection System
Data Labeling System
Methods
Time-Series Data Collection
Data Pre-Processing
Metrics for Result Evaluation
Dataset Recording Protocol
Gait Activity Recording Protocol
PD Patient Recording Protocol
Results
Recognizing Gait Events for One Subject
Recognizing Gait Events for Five Subjects
PD Patient Fog Episode's Detection
Discussion
Supplementary Materials
References
All Topics
Computer Science
Artificial Intelligence

Human Gait Activity Recognition Machine Learning Methods

Profile image of Riko SafaricRiko Safaric

Sensors

https://doi.org/10.3390/S23020745
visibility

…

description

24 pages

link

1 file

Abstract

Human gait activity recognition is an emerging field of motion analysis that can be applied in various application domains. One of the most attractive applications includes monitoring of gait disorder patients, tracking their disease progression and the modification/evaluation of drugs. This paper proposes a robust, wearable gait motion data acquisition system that allows either the classification of recorded gait data into desirable activities or the identification of common risk factors, thus enhancing the subject’s quality of life. Gait motion information was acquired using accelerometers and gyroscopes mounted on the lower limbs, where the sensors were exposed to inertial forces during gait. Additionally, leg muscle activity was measured using strain gauge sensors. As a matter of fact, we wanted to identify different gait activities within each gait recording by utilizing Machine Learning algorithms. In line with this, various Machine Learning methods were tested and compared to...

FAQs

sparkles

AI

What advanced machine learning method is most effective for gait event detection?add

The study identifies the attention-based Convolutional Neural Network and Recurrent Neural Network (CNNA + RNN) as the most reliable method for detecting specific gait events, achieving high classification metrics across varied datasets.

What unique feature does the gait motion data acquisition system offer?add

The proposed system enables semi-automatic labeling of gait data in real-time, improving the personalization and accuracy of gait analysis for individual subjects.

Which sensor placements yielded the highest performance in data collection?add

Sensor placement below the knee joint, anterior to the shin bone, resulted in the highest data acquisition performance, significantly enhancing ML classification capabilities.

How did inertial measurement units (IMUs) improve gait analysis?add

IMUs integrated into the system provided diverse data through low power, compact design, enhancing real-time accelerations measurements during various gait activities.

What challenges exist in developing universal ML algorithms for gait analysis?add

Variations in individual gait profiles and specific gait disorders like Freezing of Gait (FOG) complicate universal algorithm training, necessitating personalized ML approaches for effective detection.

Figures (22)
running for robust gait monitoring, analysis and recognition. A wearable system with sensors, a processing unit and communication, packed in a small lightweight housing is usually desired for long-term daily use. The important criteria for a wearable system are also low power consumption, sufficient external memory storage, and a user-friendly human interface for on-line data visualization and monitoring. In the case of Machine Learning (ML) methods, many of them now enable feature extraction. A combination of different learning methods can be used to recognize and extract human walking features adequately. A tradeoff between code complexity and real-time operation needs to be considered as far as implementation is concerned. The human gait is a cyclical process, with the joint operation of bones, muscles and the nervous system trying to maintain static and dynamic balance of an upright body position while in motion. The cycle can be defined as the time between two consecutive contacts of the foot striking the ground while walking and has two states, namely stance when the foot is firmly on the ground, and swing when the foot is lifted from the ground, as seen in Figure 1. Furthermore, gait phases can be divided into durations of certain events (left and right oot up/down, limb stance/swing, and limb turn) [5], and classified according to the area of nterest (sport, health, prosthetics, and daily life) [6-8], type of sensors used (floor sensors, risual, inertial or other sensors) [9,10], and last but not least, its placement (shoe, ankle, imb, and hip/back) [11,12]. The gait analysis in sports can improve walking/running echniques, prevent long-term hip, knee and ankle injuries, improve sport results, and iid the development of custom-made athlete’s textiles or sport shoes [6]. In medical pplications, a gait analysis can result in successful recognition of various gait disorders, like ‘reezing Of Gait (FOG) in Parkinson’s Disease (PD) or Cerebral Palsy (CP) patients [3]. In rosthetics applications, the gait analysis can help disabled patients after limb amputation o improve feedback in walking with a robotic prosthesis [8,13].
running for robust gait monitoring, analysis and recognition. A wearable system with sensors, a processing unit and communication, packed in a small lightweight housing is usually desired for long-term daily use. The important criteria for a wearable system are also low power consumption, sufficient external memory storage, and a user-friendly human interface for on-line data visualization and monitoring. In the case of Machine Learning (ML) methods, many of them now enable feature extraction. A combination of different learning methods can be used to recognize and extract human walking features adequately. A tradeoff between code complexity and real-time operation needs to be considered as far as implementation is concerned. The human gait is a cyclical process, with the joint operation of bones, muscles and the nervous system trying to maintain static and dynamic balance of an upright body position while in motion. The cycle can be defined as the time between two consecutive contacts of the foot striking the ground while walking and has two states, namely stance when the foot is firmly on the ground, and swing when the foot is lifted from the ground, as seen in Figure 1. Furthermore, gait phases can be divided into durations of certain events (left and right oot up/down, limb stance/swing, and limb turn) [5], and classified according to the area of nterest (sport, health, prosthetics, and daily life) [6-8], type of sensors used (floor sensors, risual, inertial or other sensors) [9,10], and last but not least, its placement (shoe, ankle, imb, and hip/back) [11,12]. The gait analysis in sports can improve walking/running echniques, prevent long-term hip, knee and ankle injuries, improve sport results, and iid the development of custom-made athlete’s textiles or sport shoes [6]. In medical pplications, a gait analysis can result in successful recognition of various gait disorders, like ‘reezing Of Gait (FOG) in Parkinson’s Disease (PD) or Cerebral Palsy (CP) patients [3]. In rosthetics applications, the gait analysis can help disabled patients after limb amputation o improve feedback in walking with a robotic prosthesis [8,13].
Figure 2. Sketch of the proposed system. The motivation of the paper is to develop a robust and lightweight system capable of human gait activity monitoring, analysis and recognition, with the aim to improve a subject’s QOL if applied in the fields of medicine, sports, or prosthetics. The proposed system (Figure 2) consists of the following components: A gait motion data acquisition system, time-series data collection, data pre-processing and data processing (classification).
Figure 2. Sketch of the proposed system. The motivation of the paper is to develop a robust and lightweight system capable of human gait activity monitoring, analysis and recognition, with the aim to improve a subject’s QOL if applied in the fields of medicine, sports, or prosthetics. The proposed system (Figure 2) consists of the following components: A gait motion data acquisition system, time-series data collection, data pre-processing and data processing (classification).
Figure 3. (a) Wearing the proposed gait motion data acquisition system; (b) close-up picture of the elastic strip equipped with sensors; (c) bare sensor unit without the elastic strip. The sensors are covered with elastic glue to increase longevity. Figure 3. (a) Wearing the proposed gait motion data acquisition system; (b) close-up picture of the
Figure 3. (a) Wearing the proposed gait motion data acquisition system; (b) close-up picture of the elastic strip equipped with sensors; (c) bare sensor unit without the elastic strip. The sensors are covered with elastic glue to increase longevity. Figure 3. (a) Wearing the proposed gait motion data acquisition system; (b) close-up picture of the
The stream of sensor data is stored into a matrix with a rate of 40 Hz (every 25 ms), where each new measurement is stored into a new matrix row. The recorded gait data matrix is visualized (Figure 5), and different gait activities can be observed throughout time. Figure 5. Recorded time-series gait data visualized (the displayed sensor information has already
The stream of sensor data is stored into a matrix with a rate of 40 Hz (every 25 ms), where each new measurement is stored into a new matrix row. The recorded gait data matrix is visualized (Figure 5), and different gait activities can be observed throughout time. Figure 5. Recorded time-series gait data visualized (the displayed sensor information has already
Table 1. Features and their characteristics.
Table 1. Features and their characteristics.
Figure 6. CWT time to frequency domain transformation visualized. (a) Presents the normalized X axis of the complementary filter; (b) visualizes the result from the CWT algorithm; (c) presents the stacked CWT result of each sensor signal. Human gait activity generally consists of frequencies between 0.5-10 Hz, therefore a frequency window of 0.2-20 Hz was selected to incorporate all the necessary information (Figure 6). CWT is a continuous 1-D transform function, so the input must be a 1-D single variable vector, from which 27 new frequency features are generated, where each feature corresponds to a specific frequency, with its value indicating its amplitude. A frequency window of 0.2-20 Hz gets divided into 27 smaller ‘frequency windows-features’. CWT was performed for all 20 acquired sensor signals independently, and later combined, generating a matrix containing 540 variables that we used to train the ML algorithms.
Figure 6. CWT time to frequency domain transformation visualized. (a) Presents the normalized X axis of the complementary filter; (b) visualizes the result from the CWT algorithm; (c) presents the stacked CWT result of each sensor signal. Human gait activity generally consists of frequencies between 0.5-10 Hz, therefore a frequency window of 0.2-20 Hz was selected to incorporate all the necessary information (Figure 6). CWT is a continuous 1-D transform function, so the input must be a 1-D single variable vector, from which 27 new frequency features are generated, where each feature corresponds to a specific frequency, with its value indicating its amplitude. A frequency window of 0.2-20 Hz gets divided into 27 smaller ‘frequency windows-features’. CWT was performed for all 20 acquired sensor signals independently, and later combined, generating a matrix containing 540 variables that we used to train the ML algorithms.
Figure 7. Auto encoder structure. the ability to learn efficient data compression and decompression. The weights and biases of the network’s neurons are adjusted to achieve minimal reconstruction loss (comparing input features and reconstructed features). The AE consists of a minimum of 3 layers: An encoder, a bottleneck and a decoder (Figure 7). The bottleneck layer includes a smaller number of neurons (the bottleneck itself), which need to learn the most vital information encompassed in the input features to reconstruct them successfully through the decoding process. Once the AE learns feature reconstruction, only the trained encoder and bottleneck layer are needed for data compression (feature reduction). The data label information was used as a ‘target’ for training the ML algorithm (a target indicating the desired output of the algorithm), so that when the trained ML algorithm is tested on an ‘unseen’ gait recording, the algorithm outputs (reveals) different gait activities within it (essentially a trained ML algorithm then creates data labels).
Figure 7. Auto encoder structure. the ability to learn efficient data compression and decompression. The weights and biases of the network’s neurons are adjusted to achieve minimal reconstruction loss (comparing input features and reconstructed features). The AE consists of a minimum of 3 layers: An encoder, a bottleneck and a decoder (Figure 7). The bottleneck layer includes a smaller number of neurons (the bottleneck itself), which need to learn the most vital information encompassed in the input features to reconstruct them successfully through the decoding process. Once the AE learns feature reconstruction, only the trained encoder and bottleneck layer are needed for data compression (feature reduction). The data label information was used as a ‘target’ for training the ML algorithm (a target indicating the desired output of the algorithm), so that when the trained ML algorithm is tested on an ‘unseen’ gait recording, the algorithm outputs (reveals) different gait activities within it (essentially a trained ML algorithm then creates data labels).
Figure 8. Visualization of the CNN + RNN algorithm architecture. A simple CNN + RNN algorithm combines the advantages of both architectures. The Convolutional layers have learnable filter parameters which can learn to produce the most meaningful feature maps for successful classification by later recurrent layers of the algorithm [35-39]. The 1D convolutional layer allows for processing of time-series data, whereas the 2D and 3D convolutional layers are designed for image processing. The CNN + RNWN algorithm architecture is presented in Figure 8. In Deep Learning, the attention mechanism learns to model the importance inside the learning data through optimizing attention weights ‘a,,’. The attention mechanism learns to enhance the most important aspects of the data, while diminishing other, less important
Figure 8. Visualization of the CNN + RNN algorithm architecture. A simple CNN + RNN algorithm combines the advantages of both architectures. The Convolutional layers have learnable filter parameters which can learn to produce the most meaningful feature maps for successful classification by later recurrent layers of the algorithm [35-39]. The 1D convolutional layer allows for processing of time-series data, whereas the 2D and 3D convolutional layers are designed for image processing. The CNN + RNWN algorithm architecture is presented in Figure 8. In Deep Learning, the attention mechanism learns to model the importance inside the learning data through optimizing attention weights ‘a,,’. The attention mechanism learns to enhance the most important aspects of the data, while diminishing other, less important
Figure 9. Visualization of the CNNA + RNN algorithm architecture.
Figure 9. Visualization of the CNNA + RNN algorithm architecture.
Table 2. Dataset activity distribution. Relatively high activity imbalances in datasets can be observed in Table 2. Inspecting the mean values of activity distributions, running was the sparsest activity (at only 6.8%), while 58.8% of the datasets consisted of walking.
Table 2. Dataset activity distribution. Relatively high activity imbalances in datasets can be observed in Table 2. Inspecting the mean values of activity distributions, running was the sparsest activity (at only 6.8%), while 58.8% of the datasets consisted of walking.
Table 3. ML algorithm parameters.
Table 3. ML algorithm parameters.
Table 3. Cont.
Table 3. Cont.
Table 4. Averaged cross-validated evaluation metrics for the corresponding nine classifiers Table 5. Standard deviation of cross-validated results. od In the remainder of this subsection, the results of the experiments are explained and presented in detail (Tables 4—7), while their further discussion is summarized in the next section. The ML performance of nine algorithms compared regarding the five evaluation metrics is presented in Table 4. The results are calculated for the CWT and RAW datasets separately. Furthermore, it is desired to use the RAW dataset for classification, since we do not have to calculate CWT at all. The best results according to each classification measure are presented as bold in the tables.
Table 4. Averaged cross-validated evaluation metrics for the corresponding nine classifiers Table 5. Standard deviation of cross-validated results. od In the remainder of this subsection, the results of the experiments are explained and presented in detail (Tables 4—7), while their further discussion is summarized in the next section. The ML performance of nine algorithms compared regarding the five evaluation metrics is presented in Table 4. The results are calculated for the CWT and RAW datasets separately. Furthermore, it is desired to use the RAW dataset for classification, since we do not have to calculate CWT at all. The best results according to each classification measure are presented as bold in the tables.
Table 6. Execution time of the classification algorithms.
Table 6. Execution time of the classification algorithms.
Table 7. Sensor importance evaluation. Table 7 presents the results from evaluating the CNN + RNN algorithm after training with different combinations of Complementary Filter (CF), Accelerometer (ACC), Gyro- scope (GYRO) and Strain Gauge (SG) RAW sensor information.
Table 7. Sensor importance evaluation. Table 7 presents the results from evaluating the CNN + RNN algorithm after training with different combinations of Complementary Filter (CF), Accelerometer (ACC), Gyro- scope (GYRO) and Strain Gauge (SG) RAW sensor information.
Table 8. RAW dataset evaluation.
Table 8. RAW dataset evaluation.
Figure 10. Visualization of the CNNA + RNN algorithm’s output on a test RAW 250 s gait record- ing; (a) Complementary filter information from ‘unseen’ gait dataset visualized; (b) illustrates the algorithm’s output on the ‘unseen’ dataset; (c) presents a comparison of the data label and algorithm’s output. Table 9. CWT dataset evaluation.
Figure 10. Visualization of the CNNA + RNN algorithm’s output on a test RAW 250 s gait record- ing; (a) Complementary filter information from ‘unseen’ gait dataset visualized; (b) illustrates the algorithm’s output on the ‘unseen’ dataset; (c) presents a comparison of the data label and algorithm’s output. Table 9. CWT dataset evaluation.
Figure 11. Visualization of the CNNA + RNN algorithm’s output on the test CWT dataset of the PD patient, that experienced five FOG episodes; (a) Complementary filter information visualized from an ‘unseen’ gait dataset; (b) illustrates the algorithm’s output on the ‘unseen’ dataset; (c) presents a comparison of the data label and the algorithm’s output. Figure 11. Visualization of the CNNA + RNN algorithm’s output on the test CWT dataset of the PD The PD patient's 24 min gait recording was split into a train dataset (60%) and test dataset (40%). The results of the latter are visualized in Figure 11a, alongside the algorithm’s output in Figure 11b. Figure 11c compares the algorithm’s output with the data label.
Figure 11. Visualization of the CNNA + RNN algorithm’s output on the test CWT dataset of the PD patient, that experienced five FOG episodes; (a) Complementary filter information visualized from an ‘unseen’ gait dataset; (b) illustrates the algorithm’s output on the ‘unseen’ dataset; (c) presents a comparison of the data label and the algorithm’s output. Figure 11. Visualization of the CNNA + RNN algorithm’s output on the test CWT dataset of the PD The PD patient's 24 min gait recording was split into a train dataset (60%) and test dataset (40%). The results of the latter are visualized in Figure 11a, alongside the algorithm’s output in Figure 11b. Figure 11c compares the algorithm’s output with the data label.
Table 10. Evaluation metrics for four classifiers.
Table 10. Evaluation metrics for four classifiers.
Table 11. Sensor importance evaluation for FOG detection. As is evident in Table 11, the CNN + RNN algorithm trained with the combined CWT data, obtained by all four sensors, showed superiority, although it also performed relatively well for some of the other tested sensor combinations, e.g., training the ML algorithm using ACC, GYRO and ACC, CF combinations produced the highest precision and sensitivity scores, respectively.
Table 11. Sensor importance evaluation for FOG detection. As is evident in Table 11, the CNN + RNN algorithm trained with the combined CWT data, obtained by all four sensors, showed superiority, although it also performed relatively well for some of the other tested sensor combinations, e.g., training the ML algorithm using ACC, GYRO and ACC, CF combinations produced the highest precision and sensitivity scores, respectively.

Related papers

The Contribution of Machine Learning in the Validation of Commercial Wearable Sensors for Gait Monitoring in Patients: A Systematic Review
Théo JOURDAN

Sensors, 2021

Gait, balance, and coordination are important in the development of chronic disease, but the ability to accurately assess these in the daily lives of patients may be limited by traditional biased assessment tools. Wearable sensors offer the possibility of minimizing the main limitations of traditional assessment tools by generating quantitative data on a regular basis, which can greatly improve the home monitoring of patients. However, these commercial sensors must be validated in this context with rigorous validation methods. This scoping review summarizes the state-of-the-art between 2010 and 2020 in terms of the use of commercial wearable devices for gait monitoring in patients. For this specific period, 10 databases were searched and 564 records were retrieved from the associated search. This scoping review included 70 studies investigating one or more wearable sensors used to automatically track patient gait in the field. The majority of studies (95%) utilized accelerometers ei...

View PDFchevron_right
Empirical Study on Human Movement Classification Using Insole Footwear Sensor System and Machine Learning
Nathan Seongheon Jeong

Sensors

This paper presents a plantar pressure sensor system (P2S2) integrated in the insoles of shoes to detect thirteen commonly used human movements including walking, stooping left and right, pulling a cart backward, squatting, descending, ascending stairs, running, and falling (front, back, right, left). Six force sensitive resistors (FSR) sensors were positioned on critical pressure points on the insoles to capture the electrical signature of pressure change in the various movements. A total of 34 adult participants were tested with the P2S2. The pressure data were collected and processed using a Principal Component Analysis (PCA) for input to the multiple machine learning (ML) algorithms, including k-NN, neural network and Support-Vector Machine (SVM) algorithms. The ML models were trained using four-fold cross-validation. Each fold kept subject data independent from other folds. The model proved effective with an accuracy of 86%, showing a promising result in predicting human moveme...

View PDFchevron_right
Wearable-Gait-Analysis-Based Activity Recognition: A Review
Stella Ansah

International Journal on Smart Sensing and Intelligent Systems

Gait analysis has been recognized as an efficient method to help realize human activity recognition; however, there is currently no existing review study focused on wearable activity recognition methods that employ gait analysis in the recognition process. In this study, different wearable-gait-analysis-based (WGA-based) activity recognition methods were summarized and compared from the aspects of wearable sensor types, data segmentation, feature extraction, and classification methods. The limitations of the current research and potential opportunities for future research in this field are also discussed.

View PDFchevron_right
Machine Learning Recognition of Gait Identity via Shoe Embedded Accelerometer
Silvia Strada

2020

Machine learning and data analytics are becoming pervasive also in the analysis of human behaviour as more and more miniaturized sensors can observe and quantify human activities. The so-called wearables are particularly tiny transducers which give the possibility to understand an individual's behaviour possibly enabling innovative services such as gait analysis based identification, foot pressure analysis, fall prevention and automatic recognition of dangerous situations. In this work we investigate an innovative experimental setting where accelerations are captured, during walking, by an Inertial Measurement Unit embedded in the shoe's sole. The goal is to identify who is wearing the shoe by simply analyzing his gait. User classification is then performed comparing different machine learning methods, relying either on the k-Neareast-Neighbour or on the Linear Discriminant Analysis algorithm. An extensive experimental campaign was carried out on five young adults and a comp...

View PDFchevron_right
Classification and Kinetic Analysis of Healthy Gait using Multiple Accelerometer Sensors
Shyam Diwakar

Procedia Computer Science, 2020

Analysis of walking-related pathologies can be done using low-cost wearable sensors that could be perceived as efficient and plausible methods for identifying gait abnormalities and neurodegenerative disorders in low economy nations. In this study, we have used mobile phone sensors that contain both accelerometers as well as gyroscopes in order to extract kinematic information and analysis of physiologically-relevant parameters to classify stance and swing among healthy volunteering subjects. Sufficient joint torque required for each limb during stance and swing phases were estimated among the subjects with weights and relevant attributes analyzed using attribute evaluators. The validation of this data was performed by pattern reconstructions in a gait simulation platform and testing whether they reproduced gait phases. Gait data categorization allowed kinematics and dynamics to be mapped to male and female subjects allowing differentiation between the genders. By translating the joint-kinematic data classification and performing torque analysis real-time allows an extension in to gait-based reconstruction of human walking and may facilitate future prosthetic devices.

View PDFchevron_right
Identification of Gait Motion Patterns Using Wearable Inertial Sensor Network
Kee Moon

Sensors, 2019

Gait signifies the walking pattern of an individual. It may be normal or abnormal, depending on the health condition of the individual. This paper considers the development of a gait sensor network system that uses a pair of wireless inertial measurement unit (IMU) sensors to monitor the gait cycle of a user. The sensor information is used for determining the normality of movement of the leg. The sensor system places the IMU sensors on one of the legs to extract the three-dimensional angular motions of the hip and knee joints while walking. The wearable sensor is custom-made at San Diego State University with wireless data transmission capability. The system enables the user to collect gait data at any site, including in a non-laboratory environment. The paper also presents the mathematical calculations to decompose movements experienced by a pair of IMUs into individual and relative three directional hip and knee joint motions. Further, a new approach of gait pattern classification...

View PDFchevron_right
Estimation of gait events and kinetic waveforms with wearable sensors and machine learning when running in an unconstrained environment
Michael Hahn

Scientific Reports

Wearable sensors and machine learning algorithms are becoming a viable alternative for biomechanical analysis outside of the laboratory. The purpose of this work was to estimate gait events from inertial measurement units (IMUs) and utilize machine learning for the estimation of ground reaction force (GRF) waveforms. Sixteen healthy runners were recruited for this study, with varied running experience. Force sensing insoles were used to measure normal foot-shoe forces, providing a proxy for vertical GRF and a standard for the identification of gait events. Three IMUs were mounted on each participant, two bilaterally on the dorsal aspect of each foot and one clipped to the back of each participant’s waistband, approximating their sacrum. Participants also wore a GPS watch to record elevation and velocity. A Bidirectional Long Short Term Memory Network (BD-LSTM) was used to estimate GRF waveforms from inertial waveforms. Gait event estimation from both IMU data and machine learning al...

View PDFchevron_right
Daily Locomotion Recognition and Prediction: A Kinematic Data-Based Machine Learning Approach
simao Carvalho

IEEE Access

More versatile, user-independent tools for recognizing and predicting locomotion modes (LMs) and LM transitions (LMTs) in natural gaits are still needed. This study tackles these challenges by proposing an automatic, user-independent recognition and prediction tool using easily wearable kinematic motion sensors for innovatively classifying several LMs (walking direction, level-ground walking, ascend and descend stairs, and ascend and descend ramps) and respective LMTs. We compared diverse state-ofthe-art feature processing and dimensionality reduction methods and machine-learning classifiers to find an effective tool for recognition and prediction of LMs and LMTs. The comparison included kinematic patterns from 10 able-bodied subjects. The more accurate tools were achieved using min-max scaling [-1;1] interval and "mRMR plus forward selection" algorithm for feature normalization and dimensionality reduction, respectively, and Gaussian support vector machine classifier. The developed tool was accurate in the recognition (accuracy > 99% and > 96%) and prediction (accuracy > 99% and > 93%) of daily LMs and LMTs, respectively, using exclusively kinematic data. The use of kinematic data yielded an effective recognition and prediction tool, predicting the LMs and LMTs one-step-ahead. This timely prediction is relevant for assistive devices providing personalized assistance in daily scenarios. The kinematic data-based machine learning tool innovatively addresses several LMs and LMTs while allowing the user to self-select the leading limb to perform LMTs, ensuring a natural gait.

View PDFchevron_right
Machine learning-based distinction of left and right foot contacts in lower back inertial sensor gait data
Andrea Cereatti

2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), 2021

Digital gait measures derived from wearable inertial sensors have been shown to support the treatment of patients with motor impairments. From a technical perspective, the detection of left and right initial foot contacts (ICs) is essential for the computation of stride-by-stride outcome measures including gait asymmetry. However, in a majority of studies only one sensor close to the center of mass is used, complicating the assignment of detected ICs to the respective foot. Therefore, we developed an algorithm including supervised machine learning (ML) models for the robust classification of left and right ICs using multiple features from the gyroscope located at the lower back. The approach was tested on a data set including 40 participants (ten healthy controls, ten hemiparetic, ten Parkinson's disease, and ten Huntington's disease patients) and reached an accuracy of 96.3% for the overall data set and up to 100.0% for the Parkinson's sub data set. These results were compared to a state-of-the-art algorithm. The ML approaches outperformed this traditional algorithm in all subgroups. Our study contributes to an improved classification of left and right ICs in inertial sensor signals recorded at the lower back and thus enables a reliable computation of clinically relevant mobility measures.

View PDFchevron_right
On Expressive Features for Gait Analysis using Lower Limb Inertial Sensor Data
Gabriele Bleser

IFAC-PapersOnLine, 2020

In this paper, we present a method to obtain explicit, expressive and interpretable gait feature signals from an inertial sensor, mounted on any segment of the lower limbs. The proposed method is invariant to the mounting orientation of the sensor, works without magnetometer information, requires no prior knowledge and can be used in real-time scenarios. Moreover, the constructed signals are robust for a wide variety of changing walking speeds and directions. We investigate the informational content of our three feature signals lying in the human sagittal plane with respect to the gait phase segmentation problem and compare them to other commonly used signals, such as the sagittal angular velocity and the norms of accelerations and angular velocities. To this end, we make use of the filter-based maximum relevance minimum redundancy algorithm, which is a classifier-independent feature selection method. For validating our approach, we consider gait data of twelve healthy subjects walking straight and in curves at self-chosen speeds with inertial sensors attached to either the thigh, shank or foot. Additionally, pressure measuring insoles are used to obtain ground truth toe-off and heel-strike gait events for reference. With those events as the gait phase transitions, the event detection is cast into a classification problem. To support the theoretical findings of the feature selection and ranking, we finally evaluate different choices of feature sets with a simple linear support vector machine classifier in an online fashion and obtain superior segmentation results with our feature signals.

View PDFchevron_right

References (57)

  1. Saunders, J.B.D.M.; Inman, V.T.; Eberhart, H.D. The Major Determinants in Normal and Pathological Gait. J. Bone Jt. Surg. 1953, 35, 543-558. [CrossRef]
  2. Bettecken, K.; Bernhard, F.; Sartor, J.; Hobert, M.A.; Hofmann, M.; Gladow, T.; van Uem, J.M.T.; Liepelt-Scarfone, I.; Maetzler, W. No relevant association of kinematic gait parameters with Health-related Quality of Life in Parkinson's disease. PLoS ONE 2017, 12, e0176816. [CrossRef] [PubMed]
  3. Mazzetta, I.; Zampogna, A.; Suppa, A.; Gumiero, A.; Pessione, M.; Irrera, F. Wearable Sensors System for an Improved Analysis of Freezing of Gait in Parkinson's Disease Using Electromyography and Inertial Signals. Sensors 2019, 19, 948. [CrossRef]
  4. Taborri, J.; Palermo, E.; Rossi, S.; Cappa, P. Gait Partitioning Methods: A Systematic Review. Sensors 2016, 16, 66. [CrossRef] [PubMed]
  5. Mo, L.; Zeng, L. Running gait pattern recognition based on cross-correlation analysis of single acceleration sensor. Math. Biosci. Eng. 2019, 16, 6242-6256. [CrossRef]
  6. DeJong, M.A.F.; Hertel, J. Validation of Foot-Strike Assessment Using Wearable Sensors During Running. J. Athl. Train. 2020, 55, 1307-1310. [CrossRef]
  7. Benson, L.C.; Clermont, C.A.; Bošnjak, E.; Ferber, R. The use of wearable devices for walking and running gait analysis outside of the lab: A systematic review. Gait Posture 2018, 63, 124-138. [CrossRef]
  8. Goršič, M.; Kamnik, R.; Ambrožič, L.; Vitiello, N.; Lefeber, D.; Pasquini, G.; Munih, M. Online Phase Detection Using Wearable Sensors for Walking with a Robotic Prosthesis. Sensors 2014, 14, 2776-2794. [CrossRef]
  9. Bejarano, N.C.; Ambrosini, E.; Pedrocchi, A.; Ferrigno, G.; Monticone, M.; Ferrante, S. A Novel Adaptive, Real-Time Algorithm to Detect Gait Events from Wearable Sensors. IEEE Trans. Neural Syst. Rehabil. Eng. 2015, 23, 413-422. [CrossRef]
  10. Jiang, X.; Chu, K.H.; Khoshnam, M.; Menon, C. A Wearable Gait Phase Detection System Based on Force Myography Techniques. Sensors 2018, 18, 1279. [CrossRef]
  11. Seel, T.; Raisch, J.; Schauer, T. IMU-Based Joint Angle Measurement for Gait Analysis. Sensors 2014, 14, 6891-6909. [CrossRef] [PubMed]
  12. Benson, L.C.; Clermont, C.A.; Watari, R.; Exley, T.; Ferber, R. Automated Accelerometer-Based Gait Event Detection during Multiple Running Conditions. Sensors 2019, 19, 1483. [CrossRef] [PubMed]
  13. Su, B.; Smith, C.; Farewik, E.G. Gait Phase Recognition Using Deep Convolutional Neural Network with Inertial Measurement Units. Biosensors 2020, 10, 109. [CrossRef]
  14. Moustakidis, S.P.; Theocharis, J.B.; Giakas, G. Subject Recognition Based on Ground Reaction Force Measurements of Gait Signals. IEEE Trans. Syst. Man Cybern. Part B (Cybernetics) 2008, 38, 1476-1485. [CrossRef]
  15. Woznowski, P.; Kaleshi, D.; Oikonomou, G.; Craddock, I. Classification and suitability of sensing technologies for activity recognition. Comput. Commun. 2016, 89-90, 34-50. [CrossRef]
  16. Aminian, K.; Najafi, B.; Büla, C.; Leyvraz, P.-F.; Robert, P. Spatio-temporal parameters of gait measured by an ambulatory system using miniature gyroscopes. J. Biomech. 2002, 35, 689-699. [CrossRef] [PubMed]
  17. Taborri, J.; Rossi, S.; Palermo, E.; Patanè, F.; Cappa, P. A Novel HMM Distributed Classifier for the Detection of Gait Phases by Means of a Wearable Inertial Sensor Network. Sensors 2014, 14, 16212-16234. [CrossRef]
  18. Sprager, S.; Juric, M.B. Inertial Sensor-Based Gait Recognition: A Review. Sensors 2015, 15, 22089-22127. [CrossRef]
  19. Benson, L.C.; Räisänen, A.M.; Clermont, C.A.; Ferber, R. Is This the Real Life, or Is This Just Laboratory? A Scoping Review of IMU-Based Running Gait Analysis. Sensors 2022, 22, 1722. [CrossRef]
  20. Lopez-Nava, I.H.; Garcia-Constantino, M.; Favela, J. Recognition of Gait Activities Using Acceleration Data from A Smartphone and A Wearable Device. Proceedings 2019, 31, 60. [CrossRef]
  21. Takeda, R.; Tadano, S.; Natorigawa, A.; Todoh, M.; Yoshinari, S. Gait posture estimation using wearable acceleration and gyro sensors. J. Biomech. 2009, 42, 2486-2494. [CrossRef] [PubMed]
  22. Prado, A.; Cao, X.; Robert, M.T.; Gordon, A.; Agrawal, S.K. Gait Segmentation of Data Collected by Instrumented Shoes Using a Recurrent Neural Network Classifier. Phys. Med. Rehabil. Clin. N. Am. 2019, 30, 355-366. [CrossRef] [PubMed]
  23. Hebenstreit, F.; Leibold, A.; Krinner, S.; Welsch, G.; Lochmann, M.; Eskofier, B.M. Effect of walking speed on gait sub phase durations. Hum. Mov. Sci. 2015, 43, 118-124. [CrossRef]
  24. Kawabata, M.; Goto, K.; Fukusaki, C.; Sasaki, K.; Hihara, E.; Mizushina, T.; Ishii, N. Acceleration patterns in the lower and upper trunk during running. J. Sports Sci. 2013, 31, 1841-1853. [CrossRef] [PubMed]
  25. Pappas, I.P.I.; Popovic, M.R.; Keller, T.; Dietz, V.; Morari, M. A reliable gait phase detection system. IEEE Trans. Neural Syst. Rehabil. Eng. 2001, 9, 113-125. [CrossRef] [PubMed]
  26. Ferreira, P.J.S.; Cardoso, J.M.P.; Mendes-Moreira, J. kNN Prototyping Schemes for Embedded Human Activity Recognition with Online Learning. Computers 2020, 9, 96. [CrossRef]
  27. Rokach, L.; Maimon, O. Data Mining with Decision Trees: Theory and Applications; Series in Machine Perception and Artificial Intelligence; WSPC: Singapore, 2014.
  28. Xu, S. Bayesian Naïve Bayes classifiers to text classification. J. Inf. Sci. 2018, 44, 48-59. [CrossRef]
  29. Ordóñez, F.J.; Roggen, D. Deep convolutional and LSTM recurrent neural networks for multimodal wearable activity recognition. Sensors 2016, 16, 115. [CrossRef]
  30. Bae, J.; Tomizuka, M. Gait phase analysis based on a Hidden Markov Model. Mechatronics 2011, 21, 961-970. [CrossRef]
  31. Yang, J.B.; Nguyen, M.N.; San, P.P.; Li, X.L.; Krishnaswamy, S. Deep Convolutional Neural Networks on Multichannel Time Series for Human Activity Recognition. In Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence, Buenos Aires, Argentina, 25-31 July 2015; pp. 3995-4001.
  32. Ben-Hur, A.; Weston, J. A User's Guide to Support Vector Machines. Methods Mol. Biol. 2009, 609, 223-239. [CrossRef]
  33. Sprager, S.; Juric, M.B. An Efficient HOS-Based Gait Authentication of Accelerometer Data. IEEE Trans. Inf. Forensics Secur. 2015, 10, 1486-1498. [CrossRef]
  34. Vaswani, A.; Shazeer, N.M.; Parmar, N.; Uszkoreit, J.; Jones, L.; Gomez, A.N.; Kaiser, L.; Polosukhin, I. Attention is All you Need. Conference on Neural Information Processing Systems (NIPS 2017) 2017, CA, USA. arXiv 2016, arXiv:1706.03762.
  35. Tajbakhsh, N.; Shin, J.Y.; Gurudu, S.R.; Hurst, R.T.; Kendall, C.B.; Gotway, M.B.; Liang, J. Convolutional Neural Networks for Medical Image Analysis: Full Training or Fine Tuning? IEEE Trans. Med. Imaging 2016, 35, 1299-1312. [CrossRef] [PubMed]
  36. Alharthi, A.S.; Yunas, S.U.; Ozanyan, K.B. Deep Learning for Monitoring of Human Gait: A Review. IEEE Sens. J. 2019, 19, 9575-9591. [CrossRef]
  37. Kiranyaz, S.; Avci, O.; Abdeljaber, O.; Ince, T.; Gabbouj, M.; Inman, D.J. 1D convolutional neural networks and applications: A survey. Mech. Syst. Signal Process 2020, 151, 107398. [CrossRef]
  38. Bisong, E. Recurrent Neural Networks (RNNs). In Building Machine Learning and Deep Learning Models on Google Cloud Platform; Apress: Berkeley, CA, USA, 2019; pp. 443-473. [CrossRef]
  39. Lempereur, M.; Rousseau, F.; Rémy-Néris, O.; Pons, C.; Houx, L.; Quellec, G.; Brochard, S. A new deep learning-based method for the detection of gait events in children with gait disorders: Proof-of-concept and concurrent validity. J. Biomech. 2020, 98, 109490.
  40. Hu, Y.; Wong, Y.; Wei, W.; Du, Y.; Kankanhalli, M.; Geng, W. A novel attention-based hybrid CNN-RNN architecture for sEMG-based gesture recognition. PLoS ONE 2018, 13, e0206049. [CrossRef]
  41. Wang, K.; He, J.; Zhang, L. Attention-Based Convolutional Neural Network for Weakly Labeled Human Activities' Recognition with Wearable Sensors. IEEE Sens. J. 2019, 19, 7598-7604. [CrossRef]
  42. Salyers, J.B.; Dong, Y.; Gai, Y. Continuous Wavelet Transform for Decoding Finger Movements from Single-Channel EEG. IEEE Trans. Biomed. Eng. 2019, 66, 1588-1597. [CrossRef]
  43. Ji, N.; Zhou, H.; Guo, K.; Samuel, O.W.; Huang, Z.; Xu, L.; Li, G. Appropriate Mother Wavelets for Continuous Gait Event Detection Based on Time-Frequency Analysis for Hemiplegic and Healthy Individuals. Sensors 2019, 19, 3462. [CrossRef]
  44. Shilane, D. Automated Feature Reduction in Machine Learning. In Proceedings of the IEEE 12th Annual Computing and Communication Workshop and Conference, Virtual, 26-29 January 2022. [CrossRef]
  45. Sun, S.-L.; Deng, Z.-L. Multi-sensor optimal information fusion Kalman filter. Automatica 2004, 40, 1017-1023. [CrossRef]
  46. Narkhede, P.; Poddar, S.; Walambe, R.; Ghinea, G.; Kotecha, K. Cascaded Complementary Filter Architecture for Sensor Fusion in Attitude Estimation. Sensors 2021, 21, 1937. [CrossRef] [PubMed]
  47. Aguiar-Conraria, L.; Soares, M.J. The continuous wavelet transform: Moving beyond uni-and bivariate analysis. J. Econ. Surv. 2014, 28, 344-375. [CrossRef]
  48. Chen, H.; Lu, F.; He, B. Topographic property of backpropagation artificial neural network: From human functional connectivity network to artificial neural network. Neurocomputing 2020, 418, 200-210. [CrossRef]
  49. Chen, T.; Xu, R.; He, Y.; Wang, X. Improving sentiment analysis via sentence type classification using BiLSTM-CRF and CNN. Expert Syst. Appl. 2017, 72, 221-230. [CrossRef]
  50. Zhao, Y.; Yang, R.; Chevalier, G.; Shah, R.C.; Romijnders, R. Applying deep bidirectional LSTM and mixture density network for basketball trajectory prediction. Optik 2017, 158, 266-272. [CrossRef]
  51. Ogawa, A.; Hori, T. Error detection and accuracy estimation in automatic speech recognition using deep bidirectional recurrent neural networks. Speech Commun. 2017, 89, 70-83. [CrossRef]
  52. Yang, J.; Bai, Y.; Lin, F.; Liu, M.; Hou, Z.; Liu, X. A novel electrocardiogram arrhythmia classification method based on stacked sparse auto-encoders and softmax regression. Int. J. Mach. Learn. Cybern. 2018, 9, 1733-1740. [CrossRef]
  53. Zhang, B.; Zhang, H.; Zhao, G.; Lian, J. Constructing a PM2.5 concentration prediction model by combining auto-encoder with Bi-LSTM neural networks. Environ. Model. Softw. 2020, 124, 104600. [CrossRef]
  54. Zhang, H.; Xiao, Z.; Wang, J.; Li, F.; Szczerbicki, E. A Novel IoT-Perceptive Human Activity Recognition (HAR) Approach Using Multihead Convolutional Attention. IEEE Internet Things J. 2019, 7, 1072-1080. [CrossRef]
  55. Hossin, M.; Sulaimn, N. A Review on Evaluation Metrics for Data Classification Evaluations. Int. J. Data Min. Knowl. Manag. Process 2015, 5, 01-11. [CrossRef]
  56. Urakami, H.; Nikaido, Y.; Kuroda, K.; Ohno, H.; Saura, R.; Okada, Y. Forward gait instability in patients with Parkinson's disease with freezing of gait. Neurosci. Res. 2021, 173, 80-89. [CrossRef] [PubMed]
  57. Disclaimer/Publisher's Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Related papers

Latest Research Trends in Gait Analysis Using Wearable Sensors and Machine Learning: A Systematic Review
Abdul Saboor

IEEE Access

Gait is the locomotion attained through the movement of limbs and gait analysis examines the patterns (normal/abnormal) depending on the gait cycle. It contributes to the development of various applications in the medical, security, sports, and fitness domains to improve the overall outcome. Among many available technologies, two emerging technologies that play a central role in modern day gait analysis are: A) wearable sensors which provide a convenient, efficient, and inexpensive way to collect data and B) Machine Learning Methods (MLMs) which enable high accuracy gait feature extraction for analysis. Given their prominent roles, this paper presents a review of the latest trends in gait analysis using wearable sensors and Machine Learning (ML). It explores the recent papers along with the publication details and key parameters such as sampling rates, MLMs, wearable sensors, number of sensors, and their locations. Furthermore, the paper provides recommendations for selecting a MLM, wearable sensor and its location for a specific application. Finally, it suggests some future directions for gait analysis and its applications.

View PDFchevron_right
A novel approach for gait analysis using activity classifier for medical analysis
Frank Vijay

Biomedical Research

Gait analysis to recognize the activity movements using wearable sensors is a cost effective, convenient, and efficient way of providing useful information for various applications in geriatrics, assessment of the performance of athletes in term of medical application and so on. In this paper, an attempt has been made to classify simple activities such as walking, running etc. In order to capture the movements, a Bluetooth Arduino microcontroller fitted with a tri-axial acceleration sensor was employed. The Arduino was programmed to record the readings of the acceleration sensor that works in collaboration with the user's smartphone or PC/laptop. In order to analyse the device performance and enable it to classify the activities correctly, Naive Bayes classifier was used. This device is fitted into the footwear of the user before the user performed the activity. Four protocols were considered for this analysis, namely walking, running, falling in the forward direction and falling in the backward direction. Five healthy adults were requested to perform this experiment. The results were consistent in the different environments considered and the device showed an overall accuracy of 90.67%. The average acceleration sensor reading in the walking scenario was found to be 0.287, 1.104 in the running scenario, 0.824 in the falling forward scenario and -0.779 in the falling backward scenario.

View PDFchevron_right
An Ambulatory Gait Monitoring System with Activity Classification and Gait Parameter Calculation Based on a Single Foot Inertial Sensor
Minsu Song

IEEE Transactions on Biomedical Engineering, 2017

For healthcare and clinical use, ambulatory gait monitoring systems using inertial sensors have been developed to estimate the user gait parameters, such as walking speed, stride time, and stride length. However, to adapt the systems effectively to daily-life activities, they need to be able to classify the gait activities of daily-life to obtain the parameters for each activity. In this study, we propose a simple classification algorithm based on a single inertial sensor for ease of use, which classifies three major gait activities: leveled walk, ramp walk, and stair walk. Method: The classification can be performed with gait parameter estimation simultaneously. The developed system that includes classification and parameter estimation algorithms was evaluated with eight healthy subjects within a gait lab and on an outdoor daily-life walking course. Results: The results showed that the estimated gait parameters were comparable to existing studies (range of walking speed root mean square error (RMSE): 0.059-0.129 m/s), and the classification accuracy was sufficiently high for all three gait activities: 98.5 % for the indoor gait lab experiment and 95.5 % for the outdoor complex daily-life walking course experiment. Conclusion: The proposed system is simple and effective for daily-life gait analysis, including gait activity classification and gait parameter estimation. for each activity.

View PDFchevron_right
Recognition of Gait Activities Using Acceleration Data from A Smartphone and A Wearable Device
Matias Garcia-Constantino

13th International Conference on Ubiquitous Computing and Ambient ‪Intelligence UCAmI 2019‬, 2019

Activity recognition is an important task in many fields, such as ambient intelligence, pervasive healthcare, and surveillance. In particular, the recognition of human gait can be useful to identify the characteristics of the places or physical spaces, such as whether the person is walking on level ground or walking down stairs in which people move. For example, ascending or descending stairs can be a risky activity for older adults because of a possible fall, which can have more severe consequences than if it occurred on a flat surface. While portable and wearable devices have been widely used to detect Activities of Daily Living (ADLs), few research works in the literature have focused on characterizing only actions of human gait. In the present study, a method for recognizing gait activities using acceleration data obtained from a smartphone and a wearable inertial sensor placed on the ankle of people is introduced. The acceleration signals were segmented based on the automatic d...

View PDFchevron_right
Using Unsupervised Machine Learning for Gait Analysis and Gait Event Detection
Abhipsha Das

2018

Automatic detection of gait events has primarily been confined to methods which require a heuristic or biometric determination of threshold values for each event, which are then stipulated as conditions while defining algorithms. This takes away from the complete automation of the process, since subject-wise threshold calculation must be done manually or using force plate data. Machine learning algorithms and neural network approaches have been proposed, but unsupervised machine learning alogrithms (K Means clustering, complete linkage) are yet to be explored and employed. Some algorithms proposed in the last two decades have presented purely automatic identification (Zeni et al., Ghoussayni et al., etc.), but the data used is either gyroscopic or captured using Vicon optical motion cameras. Microsoft Kinect is an inexpensive alternative to Vicon cameras for conducting gait analysis, and provides data for joint kinematics, especially lower limb movement data, which is pertinent to gait event detection. This project has used kinematic data collected on treadmill trials only through 3 Kinect V2s integrated using FusionKit software, and introduces machine learning techniques to perform gait event detection.

View PDFchevron_right

Related topics

  • Computer Science
  • Artificial Intelligence
  • Analytical Chemistry
  • Machine Learning
  • Gait Analysis
  • Electrical and Electronic Engine...

    • Cited by

    A Fusion-Assisted Multi-Stream Deep Learning and ESO-Controlled Newton–Raphson-Based Feature Selection Approach for Human Gait Recognition
    faiza jahangir

    Sensors

    The performance of human gait recognition (HGR) is affected by the partial obstruction of the human body caused by the limited field of view in video surveillance. The traditional method required the bounding box to recognize human gait in the video sequences accurately; however, it is a challenging and time-consuming approach. Due to important applications, such as biometrics and video surveillance, HGR has improved performance over the last half-decade. Based on the literature, the challenging covariant factors that degrade gait recognition performance include walking while wearing a coat or carrying a bag. This paper proposed a new two-stream deep learning framework for human gait recognition. The first step proposed a contrast enhancement technique based on the local and global filters information fusion. The high-boost operation is finally applied to highlight the human region in a video frame. Data augmentation is performed in the second step to increase the dimension of the p...

    View PDFchevron_right
    Artificial Intelligence Distinguishes Pathological Gait: The Analysis of Markerless Motion Capture Gait Data   Acquired by an iOS Application (TDPT-GT)
    CHIFUMI ISEKI

    Sensors

    Distinguishing pathological gait is challenging in neurology because of the difficulty of capturing total body movement and its analysis. We aimed to obtain a convenient recording with an iPhone and establish an algorithm based on deep learning. From May 2021 to November 2022 at Yamagata University Hospital, Shiga University, and Takahata Town, patients with idiopathic normal pressure hydrocephalus (n = 48), Parkinson’s disease (n = 21), and other neuromuscular diseases (n = 45) comprised the pathological gait group (n = 114), and the control group consisted of 160 healthy volunteers. iPhone application TDPT-GT captured the subjects walking in a circular path of about 1 meter in diameter, a markerless motion capture system, with an iPhone camera, which generated the three-axis 30 frames per second (fps) relative coordinates of 27 body points. A light gradient boosting machine (Light GBM) with stratified k-fold cross-validation (k = 5) was applied for gait collection for about 1 min ...

    View PDFchevron_right
    Machine Learning Classification for a Second Opinion System in the Selection of Assistive Technology in Post-Stroke Patients
    Dariusz Mikołajewski

    Applied sciences, 2023

    This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY

    View PDFchevron_right
    Gait Analysis in Neurorehabilitation: From Research to Clinical Practice
    Mirjam Bonanno

    Bioengineering

    When brain damage occurs, gait and balance are often impaired. Evaluation of the gait cycle, therefore, has a pivotal role during the rehabilitation path of subjects who suffer from neurological disorders. Gait analysis can be performed through laboratory systems, non-wearable sensors (NWS), and/or wearable sensors (WS). Using these tools, physiotherapists and neurologists have more objective measures of motion function and can plan tailored and specific gait and balance training early to achieve better outcomes and improve patients’ quality of life. However, most of these innovative tools are used for research purposes (especially the laboratory systems and NWS), although they deserve more attention in the rehabilitation field, considering their potential in improving clinical practice. In this narrative review, we aimed to summarize the most used gait analysis systems in neurological patients, shedding some light on their clinical value and implications for neurorehabilitation pra...

    View PDFchevron_right
    Bridging the Gap between Basic Research and Clinical Practice: The Growing Role of Translational Neurorehabilitation
    Mirjam Bonanno

    Medicines

    Translational neuroscience is intended as a holistic approach in the field of brain disorders, starting from the basic research of cerebral morphology and with the function of implementing it into clinical practice. This concept can be applied to the rehabilitation field to promote promising results that positively influence the patient’s quality of life. The last decades have seen great scientific and technological improvements in the field of neurorehabilitation. In this paper, we discuss the main issues related to translational neurorehabilitation, from basic research to current clinical practice, and we also suggest possible future scenarios.

    View PDFchevron_right
    580 California St., Suite 400
    San Francisco, CA, 94104
    © 2025 Academia. All rights reserved