scholarly journals Attributes’ Importance for Zero-Shot Pose-Classification Based on Wearable Sensors

Sensors ◽  
2018 ◽  
Vol 18 (8) ◽  
pp. 2485 ◽  
Author(s):  
Hiroki Ohashi ◽  
Mohammad Al-Naser ◽  
Sheraz Ahmed ◽  
Katsuyuki Nakamura ◽  
Takuto Sato ◽  
...  
Keyword(s):  
Action Recognition ◽  
Wearable Sensors ◽  
Training Data ◽  
New Method ◽  
Wearable Sensor ◽  
Baseline Method ◽  
Relative Improvement ◽  
Irrelevant Attribute ◽  
Sensor Density ◽  
Full Body

This paper presents a simple yet effective method for improving the performance of zero-shot learning (ZSL). ZSL classifies instances of unseen classes, from which no training data is available, by utilizing the attributes of the classes. Conventional ZSL methods have equally dealt with all the available attributes, but this sometimes causes misclassification. This is because an attribute that is effective for classifying instances of one class is not always effective for another class. In this case, a metric of classifying the latter class can be undesirably influenced by the irrelevant attribute. This paper solves this problem by taking the importance of each attribute for each class into account when calculating the metric. In addition to the proposal of this new method, this paper also contributes by providing a dataset for pose classification based on wearable sensors, named HDPoseDS. It contains 22 classes of poses performed by 10 subjects with 31 IMU sensors across full body. To the best of our knowledge, it is the richest wearable-sensor dataset especially in terms of sensor density, and thus it is suitable for studying zero-shot pose/action recognition. The presented method was evaluated on HDPoseDS and outperformed relative improvement of 5.9% in comparison to the best baseline method.

scholarly journals Rehabilitation and Return to Sport Assessment after Anterior Cruciate Ligament Injury: Quantifying Joint Kinematics during Complex High-Speed Tasks through Wearable Sensors

Sensors ◽  
2021 ◽  
Vol 21 (7) ◽  
pp. 2331
Author(s):  
Stefano Di Paolo ◽  
Nicola Francesco Lopomo ◽  
Francesco Della Villa ◽  
Gabriele Paolini ◽  
Giulio Figari ◽  
...  
Keyword(s):  
Anterior Cruciate Ligament ◽  
Excellent Agreement ◽  
High Speed ◽  
Cruciate Ligament ◽  
Wearable Sensors ◽  
Return To Sport ◽  
Acl Injury ◽  
Wearable Sensor ◽  
Anterior Cruciate ◽  
Full Body

The aim of the present study was to quantify joint kinematics through a wearable sensor system in multidirectional high-speed complex movements used in a protocol for rehabilitation and return to sport assessment after Anterior Cruciate Ligament (ACL) injury, and to validate it against a gold standard optoelectronic marker-based system. Thirty-four healthy athletes were evaluated through a full-body wearable sensor (MTw Awinda, Xsens) and a marker-based optoelectronic (Vicon Nexus, Vicon) system during the execution of three tasks: drop jump, forward sprint, and 90° change of direction. Clinically relevant joint angles of lower limbs and trunk were compared through Pearson’s correlation coefficient (r), and the Coefficient of Multiple Correlation (CMC). An excellent agreement (r > 0.94, CMC > 0.96) was found for knee and hip sagittal plane kinematics in all the movements. A fair-to-excellent agreement was found for frontal (r 0.55–0.96, CMC 0.63–0.96) and transverse (r 0.45–0.84, CMC 0.59–0.90) plane kinematics. Movement complexity slightly affected the agreement between the systems. The system based on wearable sensors showed fair-to-excellent concurrent validity in the evaluation of the specific joint parameters commonly used in rehabilitation and return to sport assessment after ACL injury for complex movements. The ACL professionals could benefit from full-body wearable technology in the on-field rehabilitation of athletes.

Recognizing activities of the elderly using wearable sensors: a comparison of ensemble algorithms based on boosting

Sensor Review ◽  
2019 ◽  
Vol 39 (6) ◽  
pp. 743-751 ◽  
Author(s):  
Yuchuan Wu ◽  
Shengfeng Qi ◽  
Feng Hu ◽  
Shuangbao Ma ◽  
Wen Mao ◽  
...  
Keyword(s):  
Action Recognition ◽  
Wearable Sensors ◽  
Recognition Rate ◽  
The Elderly ◽  
Human Action ◽  
Single Type ◽  
Wearable Sensor ◽  
Content Type ◽  
Linear Discriminant ◽  
Ensemble Algorithms

Purpose In human action recognition based on wearable sensors, most previous studies have focused on a single type of sensor and single classifier. This study aims to use a wearable sensor based on flexible sensors and a tri-axial accelerometer to collect action data of elderly people. It uses a statistical modeling approach based on the ensemble algorithm to classify actions and verify its validity. Design/methodology/approach Nine types of daily actions were collected by the wearable sensor device from a group of elderly volunteers, and the time-domain features of the action sequences were extracted. The dimensionality of the feature vectors was reduced by linear discriminant analysis. An ensemble learning method based on XGBoost was used to build a model of elderly action recognition. Its performance was compared with the action recognition rate of other algorithms based on the Boosting algorithm, and with the accuracy of single classifier models. Findings The effectiveness of the method was validated by three experiments. The results show that XGBoost is able to classify nine daily actions of the elderly and achieve an average recognition rate of 94.8 per cent, which is superior to single classifiers and to other ensemble algorithms. Practical implications The research could have important implications for health care, including the treatment and rehabilitation of the elderly, and the prevention of falls. Originality/value Instead of using a single type of sensor, this research used a wearable sensor to obtain daily action data of the elderly. The results show that, by using the appropriate method, the device can obtain detailed data of joint action at a low cost. Comparing differences in performance, it was concluded that XGBoost is the most suitable algorithm for building a model of elderly action recognition. This method, together with a wearable sensor, can provide key data and accurate feedback information to monitor the elderly in their rehabilitation activities.

The Feasibility and Effectiveness of Wearable Sensor Technology in the Management of Elderly Diabetics with Foot Ulcer Remission: A Proof-Of-Concept Pilot Study with Six Cases

Gerontology ◽  
2021 ◽  
pp. 1-10
Author(s):  
Chenzhen Du ◽  
Hongyan Wang ◽  
Heming Chen ◽  
Xiaoyun Fan ◽  
Dongliang Liu ◽  
...  
Keyword(s):  
Odds Ratio ◽  
Stride Length ◽  
Wearable Sensors ◽  
Foot Ulcer ◽  
Wearable Devices ◽  
Foot Ulcers ◽  
Diabetic Patients ◽  
Wearable Sensor ◽  
Double Support ◽  
Gait Parameters

Aims: Using specials wearable sensors, we explored changes in gait and balance parameters, over time, in elderly patients at high risk of diabetic foot, wearing different types of footwear. This assessed the relationship between gait and balance changes in elderly diabetic patients and the development of foot ulcers, in a bid to uncover potential benefits of wearable devices in the prognosis and management of the aforementioned complication. Methods: A wearable sensor-based monitoring system was used in middle-elderly patients with diabetes who recently recovered from neuropathic plantar foot ulcers. A total of 6 patients (age range: 55–80 years) were divided into 2 groups: the therapeutic footwear group (n = 3) and the regular footwear (n = 3) group. All subjects were assessed for gait and balance throughout the study period. Walking ability and gait pattern were assessed by allowing participants to walk normally for 1 min at habitual speed. The balance assessment program incorporated the “feet together” standing test and the instrumented modified Clinical Test of Sensory Integration and Balance. Biomechanical information was monitored at least 3 times. Results: We found significant differences in stride length (p < 0.0001), stride velocity (p < 0.0001), and double support (p < 0.0001) between the offloading footwear group (OG) and the regular footwear group on a group × time interaction. The balance test embracing eyes-open condition revealed a significant difference in Hip Sway (p = 0.004), COM Range ML (p = 0.008), and COM Position (p = 0.004) between the 2 groups. Longitudinally, the offloading group exhibited slight improvement in the performance of gait parameters over time. The stride length (odds ratio 3.54, 95% CI 1.34–9.34, p = 0.018) and velocity (odds ratio 3.13, 95% CI 1.19–8.19, p = 0.033) of OG patients increased, converse to the double-support period (odds ratio 6.20, 95% CI 1.97–19.55, p = 0.002), which decreased. Conclusions: Special wearable devices can accurately monitor gait and balance parameters in patients in real time. The finding reveals the feasibility and effectiveness of advanced wearable sensors in the prevention and management of diabetic foot ulcer and provides a solid background for future research. In addition, the development of foot ulcers in elderly diabetic patients may be associated with changes in gait parameters and the nature of footwear. Even so, larger follow-up studies are needed to validate our findings.

Improving Pattern Recognition Rate by Gaussian Hopfield Neural Network

2011 ◽  
Vol 189-193 ◽  
pp. 2042-2045 ◽  
Author(s):  
Shang Jen Chuang ◽  
Chiung Hsing Chen ◽  
Chien Chih Kao ◽  
Fang Tsung Liu
Keyword(s):  
Neural Network ◽  
Pattern Recognition ◽  
Gaussian Distribution ◽  
Test Pattern ◽  
Recognition Rate ◽  
Hopfield Neural Network ◽  
Training Data ◽  
New Method ◽  
Gaussian Filter ◽  
Testing Data

English letters cannot be recognized by the Hopfield Neural Network if it contains noise over 50%. This paper proposes a new method to improve recognition rate of the Hopfield Neural Network. To advance it, we add the Gaussian distribution feature to the Hopfield Neural Network. The Gaussian filter was added to eliminate noise and improve Hopfield Neural Network’s recognition rate. We use English letters from ‘A’ to ‘Z’ as training data. The noises from 0% to 100% were generated randomly for testing data. Initially, we use the Gaussian filter to eliminate noise and then to recognize test pattern by Hopfield Neural Network. The results are we found that if letters contain noise between 50% and 53% will become reverse phenomenon or unable recognition [6]. In this paper, we propose to uses multiple filters to improve recognition rate when letters contain noise between 50% and 53%.

GaitSense: Towards Ubiquitous Gait-Based Human Identification with Wi-Fi

2022 ◽  
Vol 18 (1) ◽  
pp. 1-24
Author(s):  
Yi Zhang ◽  
Yue Zheng ◽  
Guidong Zhang ◽  
Kun Qian ◽  
Chen Qian ◽  
...  
Keyword(s):  
Data Augmentation ◽  
Gait Recognition ◽  
Wearable Sensors ◽  
Human Identification ◽  
Training Data ◽  
Identification Accuracy ◽  
Identification System ◽  
Gait Patterns ◽  
Training Samples ◽  
Augmentation Techniques

Gait, the walking manner of a person, has been perceived as a physical and behavioral trait for human identification. Compared with cameras and wearable sensors, Wi-Fi-based gait recognition is more attractive because Wi-Fi infrastructure is almost available everywhere and is able to sense passively without the requirement of on-body devices. However, existing Wi-Fi sensing approaches impose strong assumptions of fixed user walking trajectories, sufficient training data, and identification of already known users. In this article, we present GaitSense , a Wi-Fi-based human identification system, to overcome the above unrealistic assumptions. To deal with various walking trajectories and speeds, GaitSense first extracts target specific features that best characterize gait patterns and applies novel normalization algorithms to eliminate gait irrelevant perturbation in signals. On this basis, GaitSense reduces the training efforts in new deployment scenarios by transfer learning and data augmentation techniques. GaitSense also enables a distinct feature of illegal user identification by anomaly detection, making the system readily available for real-world deployment. Our implementation and evaluation with commodity Wi-Fi devices demonstrate a consistent identification accuracy across various deployment scenarios with little training samples, pushing the limit of gait recognition with Wi-Fi signals.

scholarly journals Wearable Sensor and Internet of Things Technology for Better Medical Science: A Review

2018 ◽  
Vol 7 (4.11) ◽  
pp. 1 ◽  
Author(s):  
Arisa Olivia Putri ◽  
Musab A. M. Ali ◽  
Mohammad Saad ◽  
Sidiq Samsul Hidayat
Keyword(s):  
Internet Of Things ◽  
User Interfaces ◽  
Communication Systems ◽  
Wearable Sensors ◽  
Medical Science ◽  
Mobile User ◽  
Wearable Sensor ◽  
Main Concept ◽  
Paper Review ◽  
Internet Of Things Technology

E-health becomes one of the internet's products for healthcare. The problems of health service such as far hospital and expensive examination fees become the emergence of this technology. Consequently, people reluctant to check their health to hospital. E-health provides information on disease prevention, detecting early symptoms, and monitoring the patient's condition based on medical parameters from a far distance. Internet of things became the main concept in this system, which combines wearable sensors, communication systems, and mobile user interfaces. Reliable and valid system, easily carried, help the doctor to monitor patients from far distance expectantly to overcome the problems. The aims of this paper review are describing how an internet of things technology and wearable sensor help medical science and find the best way to create a health monitoring system.   

scholarly journals Experimental Characterization of Inkjet-Printed Stretchable Circuits for Wearable Sensor Applications

Sensors ◽  
2018 ◽  
Vol 18 (10) ◽  
pp. 3476 ◽  
Author(s):  
Jumana Abu-Khalaf ◽  
Razan Saraireh ◽  
Saleh Eisa ◽  
Ala’aldeen Al-Halhouli
Keyword(s):  
Wearable Sensors ◽  
Cost Effective ◽  
Experimental Characterization ◽  
Wearable Sensor ◽  
Successful Development ◽  
Sensor Applications ◽  
Cost Effective Method ◽  
Stretchable Circuits ◽  
Nanoparticle Ink

This paper introduces a cost-effective method for the fabrication of stretchable circuits on polydimethylsiloxane (PDMS) using inkjet printing of silver nanoparticle ink. The fabrication method, presented here, allows for the development of fully stretchable and wearable sensors. Inkjet-printed sinusoidal and horseshoe patterns are experimentally characterized in terms of the effect of their geometry on stretchability, while maintaining adequate electrical conductivity. The optimal fabricated circuit, with a horseshoe pattern at an angle of 45°, is capable of undergoing an axial stretch up to a strain of 25% with a resistance under 800 Ω. The conductivity of the circuit is fully reversible once it is returned to its pre-stretching state. The circuit could also undergo up to 3000 stretching cycles without exhibiting a significant change in its conductivity. In addition, the successful development of a novel inkjet-printed fully stretchable and wearable version of the conventional pulse oximeter is demonstrated. Finally, the resulting sensor is evaluated in comparison to its commercially available counterpart.

scholarly journals Opportunities and Challenges Surrounding the Use of Data From Wearable Sensor Devices in Health Care: Qualitative Interview Study (Preprint)

2020 ◽  
Author(s):  
Ijeoma Azodo ◽  
Robin Williams ◽  
Aziz Sheikh ◽  
Kathrin Cresswell
Keyword(s):  
Health Care ◽  
Digital Health ◽  
Wearable Sensors ◽  
Unintended Consequences ◽  
Wearable Sensor ◽  
Use Of Data ◽  
Qualitative Interview Study ◽  
Sensor Devices ◽  
Care Technology ◽  
International Experts

BACKGROUND Wearable sensors connected via networked devices have the potential to generate data that may help to automate processes of care, engage patients, and increase health care efficiency. The evidence of effectiveness of such technologies is, however, nascent and little is known about unintended consequences. OBJECTIVE Our objective was to explore the opportunities and challenges surrounding the use of data from wearable sensor devices in health care. METHODS We conducted a qualitative, theoretically informed, interview-based study to purposefully sample international experts in health care, technology, business, innovation, and social sciences, drawing on sociotechnical systems theory. We used in-depth interviews to capture perspectives on development, design, and use of data from wearable sensor devices in health care, and employed thematic analysis of interview transcripts with NVivo to facilitate coding. RESULTS We interviewed 16 experts. Although the use of data from wearable sensor devices in health and care has significant potential in improving patient engagement, there are a number of issues that stakeholders need to negotiate to realize these benefits. These issues include the current gap between data created and meaningful interpretation in health and care contexts, integration of data into health care professional decision making, negotiation of blurring lines between consumer and medical care, and pervasive monitoring of health across previously disconnected contexts. CONCLUSIONS Stakeholders need to actively negotiate existing challenges to realize the integration of data from wearable sensor devices into electronic health records. Viewing wearables as active parts of a connected digital health and care infrastructure, in which various business, personal, professional, and health system interests align, may help to achieve this.

Graphical Models for Representation and Recognition of Human Actions

2010 ◽  
pp. 31-54 ◽  
Author(s):  
Pradeep Natarajan ◽  
Ramakant Nevatia
Keyword(s):  
Action Recognition ◽  
Graphical Models ◽  
Video Processing ◽  
Hierarchical Organization ◽  
Training Data ◽  
State Transitions ◽  
Human Actions ◽  
Low Level ◽  
Spatio Temporal ◽  
Representation Scheme

Building a system for recognition of human actions from video involves two key problems - 1) designing suitable low-level features that are both efficient to extract from videos and are capable of distinguishing between events 2) developing a suitable representation scheme that can bridge the large gap between low-level features and high-level event concepts, and also handle the uncertainty and errors inherent in any low-level video processing. Graphical models provide a natural framework for representing state transitions in events and also the spatio-temporal constraints between the actors and events. Hidden Markov models(HMMs) have been widely used in several action recognition applications but the basic representation has three key deficiencies: These include unrealistic models for the duration of a sub-event, not encoding interactions among multiple agents directly and not modeling the inherent hierarchical organization of these activities. Several extensions have been proposed to address one or more of these issues and have been successfully applied in various gesture and action recognition domains. More recently, conditionalrandomfields (CRF) are becoming increasingly popular since they allow complex potential functions for modeling observations and state transitions, and also produce superior performance to HMMs when sufficient training data is available. The authors will first review the various extension of these graphical models, then present the theory of inference and learning in them and finally discuss their applications in various domains.

scholarly journals Automated P-Wave Quality Assessment for Wearable Sensors

Proceedings ◽  
2018 ◽  
Vol 4 (1) ◽  
pp. 13
Author(s):  
Diogo Tecelão ◽  
Peter Charlton
Keyword(s):  
At Risk ◽  
Quality Assessment ◽  
Wearable Sensors ◽  
P Wave ◽  
Sensor Data ◽  
Wearable Sensor ◽  
High Quality ◽  
P Waves ◽  
Ecg Signals ◽  
Wave Morphology

Hospital patients recovering from major cardiac surgery are at risk of paroxysmal atrial fibrillation (AF), an arrhythmia which can be life-threatening. Wearable sensors are routinely used for electrocardiogram (ECG) monitoring in patients at risk of AF, providing real-time AF detection. However, wearable sensors could have greater impact if used to identify the subtle changes in P-wave morphology which precede AF. This would allow prophylactic treatment to be administered, potentially preventing AF. However, ECG signals acquired by wearable sensors are susceptible to artefact, making it difficult to distinguish between physiological changes in P-wave morphology, and changes due to noise. The aim of this study was to design and assess the performance of a novel automated P-wave quality assessment tool to identify high-quality P-waves, for AF prediction. We designed a two-stage algorithm which uses P-wave template-matching to assess quality. Its performance was assessed using the AFPDB, a database of wearable sensor ECG signals acquired from both healthy subjects and patients susceptible to AF. The algorithm’s quality assessments of 97,989 P-waves were compared to manual annotations. The algorithm identified high-quality P-waves with high sensitivity (93%) and good specificity (82%), indicating that it may have utility for identifying high-quality P-waves in wearable sensor data. Measurements of P-wave morphology derived from high-quality P-waves could be used to predict AF, improving patient outcomes, and reducing healthcare costs. Further studies assessing the clinical utility of the presented tool are warranted for validation.

Sign in / Sign up

Export Citation Format

Share Document