Research

31 pages, 17550 KiB

Open AccessArticle

A Multi-Layer Classifier Model XR-KS of Human Activity Recognition for the Problem of Similar Human Activity

by Qiancheng Tan, Yonghui Qin, Rui Tang, Sixuan Wu and Jing Cao

Sensors 2023, 23(23), 9613; https://doi.org/10.3390/s23239613 - 04 Dec 2023

Viewed by 1141

Sensor-based human activity recognition is now well developed, but there are still many challenges, such as insufficient accuracy in the identification of similar activities. To overcome this issue, we collect data during similar human activities using three-axis acceleration and gyroscope sensors. We developed [...] Read more.

Sensor-based human activity recognition is now well developed, but there are still many challenges, such as insufficient accuracy in the identification of similar activities. To overcome this issue, we collect data during similar human activities using three-axis acceleration and gyroscope sensors. We developed a model capable of classifying similar activities of human behavior, and the effectiveness and generalization capabilities of this model are evaluated. Based on the standardization and normalization of data, we consider the inherent similarities of human activity behaviors by introducing the multi-layer classifier model. The first layer of the proposed model is a random forest model based on the XGBoost feature selection algorithm. In the second layer of this model, similar human activities are extracted by applying the kernel Fisher discriminant analysis (KFDA) with feature mapping. Then, the support vector machine (SVM) model is applied to classify similar human activities. Our model is experimentally evaluated, and it is also applied to four benchmark datasets: UCI DSA, UCI HAR, WISDM, and IM-WSHA. The experimental results demonstrate that the proposed approach achieves recognition accuracies of 97.69%, 97.92%, 98.12%, and 90.6%, indicating excellent recognition performance. Additionally, we performed K-fold cross-validation on the random forest model and utilized ROC curves for the SVM classifier to assess the model’s generalization ability. The results indicate that our multi-layer classifier model exhibits robust generalization capabilities. Full article

(This article belongs to the Special Issue Advances in Sensor Technologies for Wearable Applications)

► Show Figures

Figure 1

13 pages, 6168 KiB

Open AccessArticle

Red Blood Cells’ Area Deformation as the Origin of the Photoplethysmography Signal

by Lucian Evdochim, Eugen Chiriac, Marioara Avram, Lidia Dobrescu, Dragoș Dobrescu, Silviu Stanciu and Stela Halichidis

Sensors 2023, 23(23), 9515; https://doi.org/10.3390/s23239515 - 30 Nov 2023

Viewed by 628

Abstract

The origin of the photoplethysmography (PPG) signal is a debatable topic, despite plausible models being addressed. One concern revolves around the correlation between the mechanical waveform’s pulsatile nature and the associated biomechanism. The interface between these domains requires a clear mathematical or physical [...] Read more.

The origin of the photoplethysmography (PPG) signal is a debatable topic, despite plausible models being addressed. One concern revolves around the correlation between the mechanical waveform’s pulsatile nature and the associated biomechanism. The interface between these domains requires a clear mathematical or physical model that can explain physiological behavior. Describing the correct origin of the recorded optical waveform not only benefits the development of the next generation of biosensors but also defines novel health markers. In this study, the assumption of a pulsatile nature is based on the mechanism of blood microcirculation. At this level, two interconnected phenomena occur: variation in blood flow velocity through the capillary network and red blood cell (RBC) shape deformation. The latter effect was qualitatively investigated in synthetic capillaries to assess the experimental data needed for PPG model development. Erythrocytes passed through 10 µm and 6 µm microchannel widths with imposed velocities between 50 µm/s and 2000 µm/s, according to real scenarios. As a result, the length and area deformation of RBCs followed a logarithmic law function of the achieved traveling speeds. Applying radiometric expertise on top, mechanical-optical insights are obtained regarding PPG’s pulsatile nature. The mathematical equations derived from experimental data correlate microcirculation physiologic with waveform behavior at a high confidence level. The transfer function between the biomechanics and the optical signal is primarily influenced by the vasomotor state, capillary network orientation, concentration, and deformation performance of erythrocytes. Full article

(This article belongs to the Special Issue Advances in Sensor Technologies for Wearable Applications)

► Show Figures

Figure 1

11 pages, 1754 KiB

Open AccessArticle

Wheelchair Rugby Sprint Force-Velocity Modeling Using Inertial Measurement Units and Sport Specific Parameters: A Proof of Concept

by Marc Klimstra, Daniel Geneau, Melissa Lacroix, Matt Jensen, Joel Greenshields, Patrick Cormier, Ryan Brodie, Drew Commandeur and Ming-Chang Tsai

Sensors 2023, 23(17), 7489; https://doi.org/10.3390/s23177489 - 29 Aug 2023

Viewed by 682

Abstract

Background: Para-sports such as wheelchair rugby have seen increased use of inertial measurement units (IMU) to measure wheelchair mobility. The accessibility and accuracy of IMUs have enabled the quantification of many wheelchair metrics and the ability to further advance analyses such as force-velocity [...] Read more.

Background: Para-sports such as wheelchair rugby have seen increased use of inertial measurement units (IMU) to measure wheelchair mobility. The accessibility and accuracy of IMUs have enabled the quantification of many wheelchair metrics and the ability to further advance analyses such as force-velocity (FV) profiling. However, the FV modeling approach has not been refined to include wheelchair specific parameters. Purpose: The purpose of this study was to compare wheelchair rugby sprint FV profiles, developed from a wheel-mounted IMU, using current mono-exponential modeling techniques against a dynamic resistive force model with wheelchair specific resistance coefficients. Methods: Eighteen athletes from a national wheelchair rugby program performed 2 × 45 m all-out sprints on an indoor hardwood court surface. Results: Velocity modelling displayed high agreeability, with an average RMSE of 0.235 ± 0.07 m/s⁻¹ and r² of 0.946 ± 0.02. Further, the wheelchair specific resistive force model resulted in greater force and power outcomes, better aligning with previously collected measures. Conclusions: The present study highlights the proof of concept that a wheel-mounted IMU combined with wheelchair-specific FV modelling provided estimates of force and power that better account for the resistive forces encountered by wheelchair rugby athletes. Full article

(This article belongs to the Special Issue Advances in Sensor Technologies for Wearable Applications)

► Show Figures

Figure 1

25 pages, 3827 KiB

Open AccessArticle

Model-Agnostic Structural Transfer Learning for Cross-Domain Autonomous Activity Recognition

by Parastoo Alinia, Asiful Arefeen, Zhila Esna Ashari, Seyed Iman Mirzadeh and Hassan Ghasemzadeh

Sensors 2023, 23(14), 6337; https://doi.org/10.3390/s23146337 - 12 Jul 2023

Cited by 1 | Viewed by 943

Abstract

Activity recognition using data collected with smart devices such as mobile and wearable sensors has become a critical component of many emerging applications ranging from behavioral medicine to gaming. However, an unprecedented increase in the diversity of smart devices in the internet-of-things era [...] Read more.

Activity recognition using data collected with smart devices such as mobile and wearable sensors has become a critical component of many emerging applications ranging from behavioral medicine to gaming. However, an unprecedented increase in the diversity of smart devices in the internet-of-things era has limited the adoption of activity recognition models for use across different devices. This lack of cross-domain adaptation is particularly notable across sensors of different modalities where the mapping of the sensor data in the traditional feature level is highly challenging. To address this challenge, we propose ActiLabel, a combinatorial framework that learns structural similarities among the events that occur in a target domain and those of a source domain and identifies an optimal mapping between the two domains at their structural level. The structural similarities are captured through a graph model, referred to as the dependency graph, which abstracts details of activity patterns in low-level signal and feature space. The activity labels are then autonomously learned in the target domain by finding an optimal tiered mapping between the dependency graphs. We carry out an extensive set of experiments on three large datasets collected with wearable sensors involving human subjects. The results demonstrate the superiority of ActiLabel over state-of-the-art transfer learning and deep learning methods. In particular, ActiLabel outperforms such algorithms by average F1-scores of

36.3 %

,

32.7 %

, and

9.1 %

for cross-modality, cross-location, and cross-subject activity recognition, respectively. Full article

(This article belongs to the Special Issue Advances in Sensor Technologies for Wearable Applications)

► Show Figures

Figure 1

14 pages, 4208 KiB

Open AccessArticle

Dual-Sensing Piezoresponsive Foam for Dynamic and Static Loading

by Ryan A. Hanson, Cory N. Newton, Aaron Jake Merrell, Anton E. Bowden, Matthew K. Seeley, Ulrike H. Mitchell, Brian A. Mazzeo and David T. Fullwood

Sensors 2023, 23(7), 3719; https://doi.org/10.3390/s23073719 - 04 Apr 2023

Cited by 2 | Viewed by 1375

Abstract

Polymeric foams, embedded with nano-scale conductive particles, have previously been shown to display quasi-piezoelectric (QPE) properties; i.e., they produce a voltage in response to rapid deformation. This behavior has been utilized to sense impact and vibration in foam components, such as in sports [...] Read more.

Polymeric foams, embedded with nano-scale conductive particles, have previously been shown to display quasi-piezoelectric (QPE) properties; i.e., they produce a voltage in response to rapid deformation. This behavior has been utilized to sense impact and vibration in foam components, such as in sports padding and vibration-isolating pads. However, a detailed characterization of the sensing behavior has not been undertaken. Furthermore, the potential for sensing quasi-static deformation in the same material has not been explored. This paper provides new insights into these self-sensing foams by characterizing voltage response vs frequency of deformation. The correlation between temperature and voltage response is also quantified. Furthermore, a new sensing functionality is observed, in the form of a piezoresistive response to quasi-static deformation. The piezoresistive characteristics are quantified for both in-plane and through-thickness resistance configurations. The new functionality greatly enhances the potential applications for the foam, for example, as insoles that can characterize ground reaction force and pressure during dynamic and/or quasi-static circumstances, or as seat cushioning that can sense pressure and impact. Full article

(This article belongs to the Special Issue Advances in Sensor Technologies for Wearable Applications)

► Show Figures

Figure 1

15 pages, 3445 KiB

Open AccessArticle

Automatic Body Segment and Side Recognition of an Inertial Measurement Unit Sensor during Gait

by Mina Baniasad, Robin Martin, Xavier Crevoisier, Claude Pichonnaz, Fabio Becce and Kamiar Aminian

Sensors 2023, 23(7), 3587; https://doi.org/10.3390/s23073587 - 29 Mar 2023

Cited by 1 | Viewed by 1585

Abstract

Inertial measurement unit (IMU) sensors are widely used for motion analysis in sports and rehabilitation. The attachment of IMU sensors to predefined body segments and sides (left/right) is complex, time-consuming, and error-prone. Methods for solving the IMU-2-segment (I2S) pairing work properly only for [...] Read more.

Inertial measurement unit (IMU) sensors are widely used for motion analysis in sports and rehabilitation. The attachment of IMU sensors to predefined body segments and sides (left/right) is complex, time-consuming, and error-prone. Methods for solving the IMU-2-segment (I2S) pairing work properly only for a limited range of gait speeds or require a similar sensor configuration. Our goal was to propose an algorithm that works over a wide range of gait speeds with different sensor configurations while being robust to footwear type and generalizable to pathologic gait patterns. Eight IMU sensors were attached to both feet, shanks, thighs, sacrum, and trunk, and 12 healthy subjects (training dataset) and 22 patients (test dataset) with medial compartment knee osteoarthritis walked at different speeds with/without insole. First, the mean stride time was estimated and IMU signals were scaled. Using a decision tree, the body segment was recognized, followed by the side of the lower limb sensor. The accuracy and precision of the whole algorithm were 99.7% and 99.0%, respectively, for gait speeds ranging from 0.5 to 2.2 m/s. In conclusion, the proposed algorithm was robust to gait speed and footwear type and can be widely used for different sensor configurations. Full article

(This article belongs to the Special Issue Advances in Sensor Technologies for Wearable Applications)

► Show Figures

Figure 1

17 pages, 1082 KiB

Open AccessArticle

Adaptive Impedance Matching Network for Contactless Power and Data Transfer in E-Textiles

by Pim Lindeman, Annemarijn Steijlen, Jeroen Bastemeijer and Andre Bossche

Sensors 2023, 23(6), 2943; https://doi.org/10.3390/s23062943 - 08 Mar 2023

Cited by 2 | Viewed by 1475

Abstract

One of the major challenges associated with e-textiles is the connection between flexible fabric-integrated wires and rigid electronics. This work aims to increase the user experience and mechanical reliability of these connections by foregoing conventional galvanic connections in favor of inductively coupled coils. [...] Read more.

One of the major challenges associated with e-textiles is the connection between flexible fabric-integrated wires and rigid electronics. This work aims to increase the user experience and mechanical reliability of these connections by foregoing conventional galvanic connections in favor of inductively coupled coils. The new design allows for some movement between the electronics and the wires, and it relieves the mechanical strain. Two pairs of coupled coils continuously transmit power and bidirectional data across two air gaps of a few millimeters. A detailed analysis of this double inductive link and associated compensation network is presented, and the sensitivity of the network to changing conditions is explored. A proof of principle is built that demonstrates the system’s ability to self-tune based on the current–voltage phase relation. A demonstration combining 8.5 kbit/s of data transfer with a power output of 62 mW DC is presented, and the hardware is shown to support data rates of up to 240 kbit/s. This is a significant improvement of the performance of previously presented designs. Full article

(This article belongs to the Special Issue Advances in Sensor Technologies for Wearable Applications)

► Show Figures

Figure 1

20 pages, 4789 KiB

Open AccessArticle

Upper Limb Position Tracking with a Single Inertial Sensor Using Dead Reckoning Method with Drift Correction Techniques

by Lu Bai, Matthew G. Pepper, Zhibao Wang, Maurice D. Mulvenna, Raymond R. Bond, Dewar Finlay and Huiru Zheng

Sensors 2023, 23(1), 360; https://doi.org/10.3390/s23010360 - 29 Dec 2022

Cited by 1 | Viewed by 1596

Abstract

Inertial sensors are widely used in human motion monitoring. Orientation and position are the two most widely used measurements for motion monitoring. Tracking with the use of multiple inertial sensors is based on kinematic modelling which achieves a good level of accuracy when [...] Read more.

Inertial sensors are widely used in human motion monitoring. Orientation and position are the two most widely used measurements for motion monitoring. Tracking with the use of multiple inertial sensors is based on kinematic modelling which achieves a good level of accuracy when biomechanical constraints are applied. More recently, there is growing interest in tracking motion with a single inertial sensor to simplify the measurement system. The dead reckoning method is commonly used for estimating position from inertial sensors. However, significant errors are generated after applying the dead reckoning method because of the presence of sensor offsets and drift. These errors limit the feasibility of monitoring upper limb motion via a single inertial sensing system. In this paper, error correction methods are evaluated to investigate the feasibility of using a single sensor to track the movement of one upper limb segment. These include zero velocity update, wavelet analysis and high-pass filtering. The experiments were carried out using the nine-hole peg test. The results show that zero velocity update is the most effective method to correct the drift from the dead reckoning-based position tracking. If this method is used, then the use of a single inertial sensor to track the movement of a single limb segment is feasible. Full article

(This article belongs to the Special Issue Advances in Sensor Technologies for Wearable Applications)

► Show Figures

Figure 1

13 pages, 2961 KiB

Open AccessArticle

Validity of Spatio-Temporal Gait Parameters in Healthy Young Adults Using a Motion-Sensor-Based Gait Analysis System (ORPHE ANALYTICS) during Walking and Running

by Yuki Uno, Issei Ogasawara, Shoji Konda, Natsuki Yoshida, Naoki Otsuka, Yuya Kikukawa, Akira Tsujii and Ken Nakata

Sensors 2023, 23(1), 331; https://doi.org/10.3390/s23010331 - 28 Dec 2022

Cited by 2 | Viewed by 3464

Abstract

Motion sensors are widely used for gait analysis. The validity of commercial gait analysis systems is of great interest because calculating position/angle-level gait parameters potentially produces an error in the integration process of the motion sensor data; moreover, the validity of ORPHE ANALYTICS, [...] Read more.

Motion sensors are widely used for gait analysis. The validity of commercial gait analysis systems is of great interest because calculating position/angle-level gait parameters potentially produces an error in the integration process of the motion sensor data; moreover, the validity of ORPHE ANALYTICS, a motion-sensor-based gait analysis system, has not yet been examined. We examined the validity of the gait parameters calculated using ORPHE ANALYTICS relative to those calculated using conventional optical motion capture. Nine young adults performed gait tasks on a treadmill at speeds of 2–12 km/h. The three-dimensional position data and acceleration and angular velocity data of the feet were collected. The gait parameters were calculated from motion sensor data using ORPHE ANALYTICS, and optical motion capture data. Intraclass correlation coefficients [ICC_(2,1)] were calculated for relative validities. Eight items, namely, stride duration, stride length, stride frequency, stride speed, vertical height, stance phase duration, swing phase duration, and sagittal angle_IC exhibited excellent relative validities [ICC_(2,1) > 0.9]. In contrast, sagittal angle_TO and frontal angle_IC demonstrated good [ICC_(2,1) = 0.892–0.833] and moderate relative validity [ICC_(2,1) = 0.566–0.627], respectively. ORPHE ANALYTICS was found to exhibit excellent relative validities for most gait parameters. These results suggest its feasibility for gait analysis outside the laboratory setting. Full article

(This article belongs to the Special Issue Advances in Sensor Technologies for Wearable Applications)

► Show Figures

Figure 1

10 pages, 973 KiB

Open AccessArticle

Gait Variability to Phenotype Common Orthopedic Gait Impairments Using Wearable Sensors

by Junichi Kushioka, Ruopeng Sun, Wei Zhang, Amir Muaremi, Heike Leutheuser, Charles A. Odonkor and Matthew Smuck

Sensors 2022, 22(23), 9301; https://doi.org/10.3390/s22239301 - 29 Nov 2022

Cited by 3 | Viewed by 1879

Abstract

Mobility impairments are a common symptom of age-related degenerative diseases. Gait features can discriminate those with mobility disorders from healthy individuals, yet phenotyping specific pathologies remains challenging. This study aims to identify if gait parameters derived from two foot-mounted inertial measurement units (IMU) [...] Read more.

Mobility impairments are a common symptom of age-related degenerative diseases. Gait features can discriminate those with mobility disorders from healthy individuals, yet phenotyping specific pathologies remains challenging. This study aims to identify if gait parameters derived from two foot-mounted inertial measurement units (IMU) during the 6 min walk test (6MWT) can phenotype mobility impairment from different pathologies (Lumbar spinal stenosis (LSS)—neurogenic diseases, and knee osteoarthritis (KOA)—structural joint disease). Bilateral foot-mounted IMU data during the 6MWT were collected from patients with LSS and KOA and matched healthy controls (N = 30, 10 for each group). Eleven gait parameters representing four domains (pace, rhythm, asymmetry, variability) were derived for each minute of the 6MWT. In the entire 6MWT, gait parameters in all four domains distinguished between controls and both disease groups; however, the disease groups demonstrated no statistical differences, with a trend toward higher stride length variability in the LSS group (p = 0.057). Additional minute-by-minute comparisons identified stride length variability as a statistically significant marker between disease groups during the middle portion of 6WMT (3rd min: p ≤ 0.05; 4th min: p = 0.06). These findings demonstrate that gait variability measures are a potential biomarker to phenotype mobility impairment from different pathologies. Increased gait variability indicates loss of gait rhythmicity, a common feature in neurologic impairment of locomotor control, thus reflecting the underlying mechanism for the gait impairment in LSS. Findings from this work also identify the middle portion of the 6MWT as a potential window to detect subtle gait differences between individuals with different origins of gait impairment. Full article

(This article belongs to the Special Issue Advances in Sensor Technologies for Wearable Applications)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Advances in Sensor Technologies for Wearable Applications

Share This Special Issue

Special Issue Editors

Special Issue Information

Published Papers (10 papers)

Research

Further Information

Guidelines

MDPI Initiatives

Follow MDPI