scholarly journals Load Position and Weight Classification during Carrying Gait Using Wearable Inertial and Electromyographic Sensors

Sensors ◽  
2020 ◽  
Vol 20 (17) ◽  
pp. 4963
Author(s):  
Maja Goršič ◽  
Boyi Dai ◽  
Domen Novak
Keyword(s):  
Wearable Sensors ◽  
High Accuracy ◽  
Assistive Devices ◽  
Supervised Machine Learning ◽  
Classification Algorithms ◽  
Proof Of Concept ◽  
Side Load ◽  
Load Asymmetry

Lifting and carrying heavy objects is a major aspect of physically intensive jobs. Wearable sensors have previously been used to classify different ways of picking up an object, but have seen only limited use for automatic classification of load position and weight while a person is walking and carrying an object. In this proof-of-concept study, we thus used wearable inertial and electromyographic sensors for offline classification of different load positions (frontal vs. unilateral vs. bilateral side loads) and weights during gait. Ten participants performed 19 different carrying trials each while wearing the sensors, and data from these trials were used to train and evaluate classification algorithms based on supervised machine learning. The algorithms differentiated between frontal and other loads (side/none) with an accuracy of 100%, between frontal vs. unilateral side load vs. bilateral side load with an accuracy of 96.1%, and between different load asymmetry levels with accuracies of 75–79%. While the study is limited by a lack of electromyographic sensors on the arms and a limited number of load positions/weights, it shows that wearable sensors can differentiate between different load positions and weights during gait with high accuracy. In the future, such approaches could be used to control assistive devices or for long-term worker monitoring in physically demanding occupations.

scholarly journals Development and Validation of Open-Source Activity Intensity Count and Activity Intensity Classification Algorithms from Raw Acceleration Signals of Wearable Sensors

Sensors ◽  
2020 ◽  
Vol 20 (23) ◽  
pp. 6767
Author(s):  
Isabelle Poitras ◽  
Jade Clouâtre ◽  
Laurent J. Bouyer ◽  
François Routhier ◽  
Catherine Mercier ◽  
...  
Keyword(s):  
Open Source ◽  
Wearable Sensors ◽  
Classification Algorithms ◽  
Classification Methods ◽  
Commercial Activity ◽  
Activity Intensity ◽  
Development And Validation ◽  
Intensity Classification ◽  
Acceleration Signals

Background: A popular outcome in rehabilitation studies is the activity intensity count, which is typically measured from commercially available accelerometers. However, the algorithms are not openly available, which impairs long-term follow-ups and restricts the potential to adapt the algorithms for pathological populations. The objectives of this research are to design and validate open-source algorithms for activity intensity quantification and classification. Methods: Two versions of a quantification algorithm are proposed (fixed [FB] and modifiable bandwidth [MB]) along with two versions of a classification algorithm (discrete [DM] vs. continuous methods [CM]). The results of these algorithms were compared to those of a commercial activity intensity count solution (ActiLife) with datasets from four activities (n = 24 participants). Results: The FB and MB algorithms gave similar results as ActiLife (r > 0.96). The DM algorithm is similar to a ActiLife (r ≥ 0.99). The CM algorithm differs (r ≥ 0.89) but is more precise. Conclusion: The combination of the FB algorithm with the DM results is a solution close to that of ActiLife. However, the MB version remains valid while being more adaptable, and the CM is more precise. This paper proposes an open-source alternative for rehabilitation that is compatible with several wearable devices and not dependent on manufacturer commercial decisions.

scholarly journals Supervised Methods Applied to the Construction of a Vision System for the Classification of Cocoa Beans in the Cut-Test

2019 ◽  
Author(s):  
Felipe Antunes ◽  
Anne Canuto ◽  
Benjamin Bedregal ◽  
Eduardo Palmeira ◽  
Iaslan Silva
Keyword(s):  
System Performance ◽  
Vision System ◽  
Attribute Selection ◽  
Supervised Machine Learning ◽  
Classification Algorithms ◽  
Cocoa Beans ◽  
Machine Learning Methods ◽  
Supervised Methods ◽  
Classification Tasks

Supervised machine learning methods, also known as classification algorithms, have been widely used in the literature for many classification tasks. In this context, some aspects of these algorithms, as the used attributes used and the form they were built, have a direct impact in the system performance. Therefore, in this paper, we evaluate the application of classification algorithms, along with attribute selection, to propose an improved version of a vision system that performs the classification of cocoa beans. The main aim of this investigation is to improve the performance of a cocoa classification system that aims at helping farmers to classify the different cocoa beans based on images of these beans.

scholarly journals An evaluation of machine learning classifiers for next-generation, continuous-ethogram smart trackers

2021 ◽  
Vol 9 (1) ◽  
Author(s):  
Hui Yu ◽  
Jian Deng ◽  
Ran Nathan ◽  
Max Kröschel ◽  
Sasha Pekarsky ◽  
...  
Keyword(s):  
Machine Learning ◽  
Feature Reduction ◽  
Supervised Machine Learning ◽  
Support Vector ◽  
Learning Methods ◽  
Machine Learning Methods ◽  
Behavioural Research ◽  
Intermittent Sampling

Abstract Background Our understanding of movement patterns and behaviours of wildlife has advanced greatly through the use of improved tracking technologies, including application of accelerometry (ACC) across a wide range of taxa. However, most ACC studies either use intermittent sampling that hinders continuity or continuous data logging relying on tracker retrieval for data downloading which is not applicable for long term study. To allow long-term, fine-scale behavioural research, we evaluated a range of machine learning methods for their suitability for continuous on-board classification of ACC data into behaviour categories prior to data transmission. Methods We tested six supervised machine learning methods, including linear discriminant analysis (LDA), decision tree (DT), support vector machine (SVM), artificial neural network (ANN), random forest (RF) and extreme gradient boosting (XGBoost) to classify behaviour using ACC data from three bird species (white stork Ciconia ciconia, griffon vulture Gyps fulvus and common crane Grus grus) and two mammals (dairy cow Bos taurus and roe deer Capreolus capreolus). Results Using a range of quality criteria, SVM, ANN, RF and XGBoost performed well in determining behaviour from ACC data and their good performance appeared little affected when greatly reducing the number of input features for model training. On-board runtime and storage-requirement tests showed that notably ANN, RF and XGBoost would make suitable on-board classifiers. Conclusions Our identification of using feature reduction in combination with ANN, RF and XGBoost as suitable methods for on-board behavioural classification of continuous ACC data has considerable potential to benefit movement ecology and behavioural research, wildlife conservation and livestock husbandry.

scholarly journals Semi-automated regional classification of the style of activity of slow rock slope deformations using PS InSAR and SqueeSAR velocity data

2021 ◽  
Author(s):  
Chiara Crippa ◽  
Elena Valbuzzi ◽  
Paolo Frattini ◽  
Giovanni B. Crosta ◽  
Margherita C. Spreafico ◽  
...  
Keyword(s):  
Rock Slope ◽  
Supervised Machine Learning ◽  
Displacement Rate ◽  
Persistent Scatterer Interferometry ◽  
Internal Damage ◽  
Machine Learning Classification ◽  
Mountainous Regions ◽  
Definition Of

<p>Large slow rock-slope deformations are widespread in alpine environments and mountainous regions worldwide. They evolve over long time by progressive failure processes, resulting in slow movements that impact infrastructures and can eventually evolve into catastrophic rockslides. A robust characterization of the activity of these phenomena is thus required to cope with their long-term threats.</p><p>Displacement rates measured by remote sensing and ground-based techniques only provide a snapshot of long-term, variable trends of activity and are insufficient to capture the behavior of slow rock slope deformations in a long-term risk management perspective. We thus propose to adopt a more complete approach based on a re-definition of “style of activity”, including displacement rate, segmentation/heterogeneity, kinematics, internal damage and accumulated strain. To this aim, we developed a novel approach combining persistent-scatterer interferometry (PSI) and systematic geomorphological mapping, to obtain an objective semi-automated characterization and classification of 208 slow rock slope deformations in Lombardia (Italian Central Alps). Through a peak analysis of displacement rate distributions we characterized the degree of internal segmentation of mapped slow rock slope deformations and highlighted the presence of nested sectors with differential activity. Then, we used an original approach to automatically characterize the kinematics of each landslide (translational, compound, or rotational) by combining a 2DInSAR velocity vector decomposition and a supervised machine learning classification. Finally, we combined Principal Component and K-medoid Cluster multivariate statistical analyses to classify slow rock slope deformations into groups with consistent styles of activity. We classified DSGSDs and large landslides respectively in five and two representative groups described by different degree of internal segmentation and kinematics that significant influence the evolutionary behavior and affect the definition of representative displacement rates. Our results provide a statistical evidence that phenomena classified as “Deep-Seated Gravitational Slope deformations” (DSGSD) and “large landslides” actually have different mechanisms and/or evolutionary stages, mirrored by different morphological features that testify higher accumulated internal deformation for large landslides with respect to DSGSDs. Our statistical classification of rock-slope deformation style of activity further highlighted the different risk potentials associated to each one of the seven descriptive groups in a practical perspective, taking into account the most significant parameters (rate, volume and heterogeneity) to assess risks related to the interaction between slow movements and sensitive elements.</p><p>Our analysis benefits from both deterministic and statistical components to perform a complete regional screening of slow rock slope deformations and to prioritize site-specific, engineering geological analyses of critical slopes depending on the most important factors conditioning their long-term style of activity. Our methodology is readily applicable to different datasets and provides an objective and cost-effective support to land planning and the prioritization of local-scale studies aimed at granting safety and infrastructure integrity.</p>

scholarly journals Analysis of Data for Diabetics Patient

2017 ◽  
Vol 13 (15) ◽  
pp. 216
Author(s):  
Korobi Saha Koli ◽  
Sajjad Waheed
Keyword(s):  
Heart Attack ◽  
Diabetes Patient ◽  
High Accuracy ◽  
The Body ◽  
Classification Algorithms ◽  
Hoeffding Tree ◽  
Endocrine Hormone ◽  
Sugar Levels ◽  
Hormone System

Diabetes, a disease responsible for different kinds of diseases such as heart attack, kidney disease, blindness and renal failure etc. The most common disorder is the endocrine (hormone) system, occurs when blood sugar levels in the body consistently stay above normal. There are two types of diabetic; one is body's inability to make insulin and another is body not responding to the effects of insulin. In our developing country Bangladesh, Diabetes is a costly disease whose risk is increasing at alarming rate. This paper evaluates the selected classification algorithms for the classification of some Diabetes patient datasets. Classification algorithms considered here are Naive Bayes classification (NBC), Bagging algorithm, KStar algorithm, Logistic algorithm and Hoeffding tree. These algorithms are evaluated based on four criteria: Accuracy, Precision, Sensitivity and Specificity. Collected datasets of diabetes affected people are firstly preprocessed then some investigation based on mentioned algorithm has been executed successfully. From the investigation result it is found that, KStar algorithm is the best as it gives high accuracy with the low error. Here it is said that, some parameters are responsible for diabetes.

A Pressure Sores Assessment System for Diagnosis and Decision Making: The Application of Convolutional Neuron Networks (Preprint)

2020 ◽  
Author(s):  
Tom J. Liu ◽  
Yuan-Chia Chu ◽  
Christian Mesakh ◽  
Yu-Chun Chen ◽  
Che-Wei Chang ◽  
...  
Keyword(s):  
Deep Learning ◽  
Soft Tissues ◽  
Assessment System ◽  
University Hospital ◽  
Supervised Machine Learning ◽  
Diagnostic Model ◽  
Pressure Sores ◽  
Term Care

BACKGROUND Pressure sores are a common problem in hospital care and long-term care. Pressure sores are caused by prolonged compression of soft tissues, which can cause local tissue damage and even lead to serious infections. Pressure sores can result in poor prognosis, long-term hospitalization, and increased medical costs, which are especially problematic in an aging society. OBJECTIVE This study uses deep learning to diagnose pressure sores and assist in making treatment decisions, thus providing additional reference for first-line caregivers. METHODS We utilized retrospective research of medical records to find photos of patients with pressure sores at National Taiwan University Hospital from 2016 to 2019. We removed the photos which were vague, underexposed, or overexposed, and then labeled the remaining photos as “infected” or “uninfected” and “extensive necrosis”, “moderate necrosis” or “limited necrosis”. Supervised machine learning was then used, and Convolutional Neural Networks (CNNs) were applied for deep learning to construct a diagnostic model. Finally, we tested the constructed model with these photos to verify its accuracy. RESULTS For the task of classification of infected and non-infected wounds, our CNN model achieved an accuracy of about 98%. For the task of classification of necrotic tissues, our model achieved accuracy of about 94%. CONCLUSIONS Compared with traditional algorithms, our deep learning model achieved higher accuracy, making it applicable in clinical circumstances.

Optimization of K Value in KNN Algorithm for Spam and Ham Email Classification

2020 ◽  
Vol 4 (2) ◽  
pp. 377-383
Author(s):  
Eko Laksono ◽  
Achmad Basuki ◽  
Fitra Bachtiar
Keyword(s):  
Frequency Distribution ◽  
Confusion Matrix ◽  
High Accuracy ◽  
K Value ◽  
A Value ◽  
Optimal Value ◽  
Classification Evaluation ◽  
Email Spam ◽  
Email Classification

There are many cases of email abuse that have the potential to harm others. This email abuse is commonly known as spam, which contains advertisements, phishing scams, and even malware. This study purpose to know the classification of email spam with ham using the KNN method as an effort to reduce the amount of spam. KNN can classify spam or ham in an email by checking it using a different K value approach. The results of the classification evaluation using confusion matrix resulted in the KNN method with a value of K = 1 having the highest accuracy value of 91.4%. From the results of the study, it is known that the optimization of the K value in KNN using frequency distribution clustering can produce high accuracy of 100%, while k-means clustering produces an accuracy of 99%. So based on the results of the existing accuracy values, the frequency distribution clustering and k-means clustering can be used to optimize the K-optimal value of the KNN in the classification of existing spam emails.

MODERN PROBLEMATICS AND ANALYSIS OF EFFICIENCY OF APPLICATION OF DIFFERENT APPROACHES TO DESTRUCTION OF INFORMATION ON MAGNETIC MEDIA

2019 ◽  
Vol 2 (12) ◽  
pp. 154-164
Author(s):  
O. Semenenko ◽  
O. Vodchyts ◽  
V. Koverga ◽  
R. Lukash ◽  
O. Lutsenko
Keyword(s):  
Armed Forces ◽  
Information Storage ◽  
Storage Device ◽  
Magnetic Media ◽  
Long Term Storage ◽  
Physical Impact ◽  
And Storage ◽  
Term Storage

The introduction and active use of information transmission and storage systems in the Ministry of Defense (MoD) of Ukraine form the need to develop ways of guaranteed removal of data from media after their use or long-term storage. Such a task is an essential component of the functioning of any information security system. The article analyzes the problems of guaranteed destruction of information on magnetic media. An overview of approaches to the guaranteed destruction of information on magnetic media of different types is presented, and partial estimates of the effectiveness of their application are given by some generally accepted indicators of performance evaluation. The article also describes the classification of methods of destruction of information depending on the influence on its medium. The results of the analysis revealed the main problems of application of software methods and methods of demagnetization of the information carrier. The issue of guaranteed destruction of information from modern SSD devices, which are actively used in the formation of new systems of information accumulation and processing, became particularly relevant in the article. In today's conditions of development of the Armed Forces of Ukraine, methods of mechanical and thermal destruction are more commonly used today. In the medium term, the vector of the use of information elimination methods will change towards the methods of physical impact by the pulsed magnetic field and the software methods that allow to store the information storage device, but this today requires specialists to develop new ways of protecting information in order to avoid its leakage.

Influence of internal and external risks on sustainability of financial system of Аltai krai

2018 ◽  
Vol 35 (4) ◽  
pp. 133-136
Author(s):  
R. N. Ibragimov
Keyword(s):  
Sustainable Development ◽  
Financial Stability ◽  
Management Strategy ◽  
Financial System ◽  
Risk Management Strategy ◽  
The Sustainable Development ◽  
The Stability ◽  
The Impact

The article examines the impact of internal and external risks on the stability of the financial system of the Altai Territory. Classification of internal and external risks of decline, affecting the sustainable development of the financial system, is presented. A risk management strategy is proposed that will allow monitoring of risks, thereby these measures will help reduce the loss of financial stability and ensure the long-term development of the economy of the region.

Sign in / Sign up

Export Citation Format

Share Document