Human Activity Recognition by Wearable Sensors : Comparison of different classifiers for real-time applications

Leonardis, G; Rosati, Samanta; Balestra, Gabriella; Agostini, Valentina; Panero, Elisa; Gastaldi, Laura; Knaflitz, Marco

doi:10.1109/memea.2018.8438750

Cited by 68 publications

(42 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The feature extraction methods of the HAR system can be divided into three categories, namely, time features, frequency features, and a combination of both [9]. Jarraya et al selected 280 features from a total of 561 by means of a nonlinear Choquet integral feature selection approach, classified six basic actions by using the random forest, and finally obtained a better classification effect.…”

Section: Introductionmentioning

confidence: 99%

A Feature Selection and Classification Method for Activity Recognition Based on an Inertial Sensing Unit

Fan

Jia

2019

Information

View full text Add to dashboard Cite

The purpose of activity recognition is to identify activities through a series of observations of the experimenter’s behavior and the environmental conditions. In this study, through feature selection algorithms, we researched the effects of a large number of features on human activity recognition (HAR) assisted by an inertial measurement unit (IMU), and applied them to smartphones of the future. In the research process, we considered 585 features (calculated from tri-axial accelerometer and tri-axial gyroscope data). We comprehensively analyzed the features of signals and classification methods. Three feature selection algorithms were considered, and the combination effect between the features was used to select a feature set with a significant effect on the classification of the activity, which reduced the complexity of the classifier and improved the classification accuracy. We used five classification methods (support vector machine [SVM], decision tree, linear regression, Gaussian process, and threshold selection) to verify the classification accuracy. The activity recognition method we proposed could recognize six basic activities (BAs) (standing, going upstairs, going downstairs, walking, lying, and sitting) and postural transitions (PTs) (stand-to-sit, sit-to-stand, stand-to-lie, lie-to-stand, sit-to-lie, and lie-to-sit), with an average accuracy of 96.4%.

show abstract

Section: Introductionmentioning

confidence: 99%

A Feature Selection and Classification Method for Activity Recognition Based on an Inertial Sensing Unit

Fan

Jia

2019

Information

View full text Add to dashboard Cite

show abstract

“…This architecture was deemed sufficient after the initial segmentation based on laboratory experiments and an analysis of the literature. As seen in Reference [ 36 ], an MLP implemented with 1 hidden layer, 20 neurons, 8 output classes and a sigmoid transfer function accomplishes a

accuracy per subject. The data set consists of 15 healthy subjects and 8 different human activities.…”

Section: Materials and Methodsmentioning

confidence: 99%

Wearable Sensor Network for Biomechanical Overload Assessment in Manual Material Handling

Giannini

Bassani

Avizzano

et al. 2020

Sensors

View full text Add to dashboard Cite

The assessment of risks due to biomechanical overload in manual material handling is nowadays mainly based on observational methods in which an expert rater visually inspects videos of the working activity. Currently available sensing wearable technologies for motion and muscular activity capture enables to advance the risk assessment by providing reliable, repeatable, and objective measures. However, existing solutions do not address either a full body assessment or the inclusion of measures for the evaluation of the effort. This article proposes a novel system for the assessment of biomechanical overload, capable of covering all areas of ISO 11228, that uses a sensor network composed of inertial measurement units (IMU) and electromyography (EMG) sensors. The proposed method is capable of gathering and processing data from three IMU-based motion capture systems and two EMG capture devices. Data are processed to provide both segmentation of the activity and ergonomic risk score according to the methods reported in the ISO 11228 and the TR 12295. The system has been tested on a challenging outdoor scenario such as lift-on/lift-off of containers on a cargo ship. A comparison of the traditional evaluation method and the proposed one shows the consistency of the proposed system, its time effectiveness, and its potential for deeper analyses that include intra-subject and inter-subjects variability as well as a quantitative biomechanical analysis.

show abstract

“…To perform feature selection we used genetic algorithms (GAs), that is, heuristic algorithms belonging to the computational intelligence field. 25,26 Each GA solution was coded as a binary 24 bits vector, one bit representing a feature: a "0" in a given position identified a feature not selected whereas a "1" labeled a feature included in the final subset. Each solution of the GA was evaluated by a fitness function that measured the ability of the corresponding feature subset to obtain a Gaussian naïve Bayesian classifier able to classify metastases of the training set.…”

Section: Feature Selection and Radiomics Model Developmentmentioning

confidence: 99%

Radiomics predicts response of individual HER2‐amplified colorectal cancer liver metastases in patients treated with HER2‐targeted therapy

Giannini

Rosati

Defeudis

et al. 2020

Intl Journal of Cancer

View full text Add to dashboard Cite

The aim of our study was to develop and validate a machine learning algorithm to predict response of individual HER2-amplified colorectal cancer liver metastases (lmCRC) undergoing dual HER2-targeted therapy. Twenty-four radiomics features were extracted after 3D manual segmentation of 141 lmCRC on pretreatment portal CT scans of a cohort including 38 HER2-amplified patients; feature selection was then performed using genetic algorithms. lmCRC were classified as nonresponders (R−), if their largest diameter increased more than 10% at a CT scan performed after 3 months of treatment, responders (R+) otherwise. Sensitivity, specificity, negative (NPV) and positive (PPV) predictive values in correctly classifying individual lesion and overall patient response were assessed on a training dataset and then validated on a second dataset using a Gaussian naïve Bayesian classifier. Per-lesion sensitivity, specificity, NPV and PPV were 89%, 85%, 93%, 78% and 90%, 42%, 73%, 71% respectively in the testing and validation datasets. Per-patient sensitivity and specificity were 92% and 86%. Heterogeneous response was observed in 9 of 38 patients (24%). Five of nine patients were carriers of nonresponder lesions correctly classified as such by our radiomics signature, including four of seven harboring only one nonresponder lesion. The developed method has been proven effective in predicting behavior of individual metastases to targeted treatment in a cohort of HER2 amplified patients. The model accurately detects responder lesions and identifies nonresponder lesions in patients with heterogeneous response, potentially paving the way to multimodal treatment in selected patients. Further validation will be needed to confirm our findings.

show abstract

Human Activity Recognition by Wearable Sensors : Comparison of different classifiers for real-time applications

Cited by 68 publications

References 18 publications

A Feature Selection and Classification Method for Activity Recognition Based on an Inertial Sensing Unit

A Feature Selection and Classification Method for Activity Recognition Based on an Inertial Sensing Unit

Wearable Sensor Network for Biomechanical Overload Assessment in Manual Material Handling

Radiomics predicts response of individual HER2‐amplified colorectal cancer liver metastases in patients treated with HER2‐targeted therapy

Contact Info

Product

Resources

About