Classification of Human Daily Activities Using Ensemble Methods Based on Smartphone Inertial Sensors

TCN-attention-HAR: human activity recognition based on attention mechanism time convolutional network

Wearable sensors are widely used in medical applications and human-computer interaction because of their portability and powerful privacy. Human activity identification based on sensor data plays a vital role in these fields. Therefore, it is important to improve the recognition performance of different types of actions. Aiming at the problems of insufficient time-varying feature extraction and gradient explosion caused by too many network layers, a time convolution network recognition model with attention mechanism (TCN-Attention-HAR) was proposed. The model effectively recognizes and emphasizes the key feature information. The ability of extracting temporal features from TCN (temporal convolution network) is improved by using the appropriate size of the receiver domain. In addition, attention mechanisms are used to assign higher weights to important information, enabling models to learn and identify human activities more effectively. The performance of the Open Data Set (WISDM, PAMAP2 and USC-HAD) is improved by 1.13%, 1.83% and 0.51%, respectively, compared with other advanced models, these results clearly show that the network model presented in this paper has excellent recognition performance. In the knowledge distillation experiment, the parameters of student model are only about 0.1% of those of teacher model, and the accuracy of the model has been greatly improved, and in the WISDM data set, compared with the teacher's model, the accuracy is 0.14% higher.

A Machine Learning Approach for Walking Classification in Elderly People with Gait Disorders

Walking ability of elderly individuals, who suffer from walking difficulties, is limited, which restricts their mobility independence. The physical health and well-being of the elderly population are affected by their level of physical activity. Therefore, monitoring daily activities can help improve the quality of life. This becomes especially a huge challenge for those, who suffer from dementia and Alzheimer's disease. Thus, it is of great importance for personnel in care homes/rehabilitation centers to monitor their daily activities and progress. Unlike normal subjects, it is required to place the sensor on the back of this group of patients, which makes it even more challenging to detect walking from other activities. With the latest advancements in the field of health sensing and sensor technology, a huge amount of accelerometer data can be easily collected. In this study, a Machine Learning (ML) based algorithm was developed to analyze the accelerometer data collected from patients with walking difficulties, who live in one of the municipalities in Denmark. The ML algorithm is capable of accurately classifying the walking activity of these individuals with different walking abnormalities. Various statistical, temporal, and spectral features were extracted from the time series data collected using an accelerometer sensor placed on the back of the participants. The back sensor placement is desirable in patients with dementia and Alzheimer's disease since they may remove visible sensors to them due to the nature of their diseases. Then, an evolutionary optimization algorithm called Particle Swarm Optimization (PSO) was used to select a subset of features to be used in the classification step. Four different ML classifiers such as k-Nearest Neighbors (kNN), Random Forest (RF), Stacking Classifier (Stack), and Extreme Gradient Boosting (XGB) were trained and compared on an accelerometry dataset consisting of 20 participants. These models were evaluated using the leave-one-group-out cross-validation (LOGO-CV) technique. The Stack model achieved the best performance with average sensitivity, positive predictive values (precision), F₁-score, and accuracy of 86.85%, 93.25%, 88.81%, and 93.32%, respectively, to classify walking episodes. In general, the empirical results confirmed that the proposed models are capable of classifying the walking episodes despite the challenging sensor placement on the back of the patients, who suffer from walking disabilities.

An Efficient Prediction System for Coronary Heart Disease Risk Using Selected Principal Components and Hyperparameter Optimization

Medical science-related studies have reinforced that the prevalence of coronary heart disease which is associated with the heart and blood vessels has been the most significant cause of health loss and death globally. Recently, data mining and machine learning have been used to detect diseases based on the unique characteristics of a person. However, these techniques have often posed challenges due to the complexity in understanding the objective of the datasets, the existence of too many factors to analyze as well as lack of performance accuracy. This research work is of two-fold effort: firstly, feature extraction and selection. This entails extraction of the principal components, and consequently, the Correlation-based Feature Selection (CFS) method was applied to select the finest principal components of the combined (Cleveland and Statlog) heart dataset. Secondly, by applying datasets to three single and three ensemble classifiers, the best hyperparameters that reflect the pre-eminent predictive outcomes were investigated. The experimental result reveals that hyperparameter optimization has improved the accuracy of all the models. In the comparative studies, the proposed work outperformed related works with an accuracy of 97.91%, and an AUC of 0.996 by employing six optimal principal components selected from the CFS method and optimizing parameters of the Rotation Forest ensemble classifier.

Top-Down System for Multi-Person 3D Absolute Pose Estimation from Monocular Videos

Two-dimensional (2D) multi-person pose estimation and three-dimensional (3D) root-relative pose estimation from a monocular RGB camera have made significant progress recently. Yet, real-world applications require depth estimations and the ability to determine the distances between people in a scene. Therefore, it is necessary to recover the 3D absolute poses of several people. However, this is still a challenge when using cameras from single points of view. Furthermore, the previously proposed systems typically required a significant amount of resources and memory. To overcome these restrictions, we herein propose a real-time framework for multi-person 3D absolute pose estimation from a monocular camera, which integrates a human detector, a 2D pose estimator, a 3D root-relative pose reconstructor, and a root depth estimator in a top-down manner. The proposed system, called Root-GAST-Net, is based on modified versions of GAST-Net and RootNet networks. The efficiency of the proposed Root-GAST-Net system is demonstrated through quantitative and qualitative evaluations on two benchmark datasets, Human3.6M and MuPoTS-3D. On all evaluated metrics, our experimental results on the MuPoTS-3D dataset outperform the current state-of-the-art by a significant margin, and can run in real-time at 15 fps on the Nvidia GeForce GTX 1080.

Opportunities for Smartphone Sensing in E-Health Research: A Narrative Review

Recent years have seen significant advances in the sensing capabilities of smartphones, enabling them to collect rich contextual information such as location, device usage, and human activity at a given point in time. Combined with widespread user adoption and the ability to gather user data remotely, smartphone-based sensing has become an appealing choice for health research. Numerous studies over the years have demonstrated the promise of using smartphone-based sensing to monitor a range of health conditions, particularly mental health conditions. However, as research is progressing to develop the predictive capabilities of smartphones, it becomes even more crucial to fully understand the capabilities and limitations of using this technology, given its potential impact on human health. To this end, this paper presents a narrative review of smartphone-sensing literature from the past 5 years, to highlight the opportunities and challenges of this approach in healthcare. It provides an overview of the type of health conditions studied, the types of data collected, tools used, and the challenges encountered in using smartphones for healthcare studies, which aims to serve as a guide for researchers wishing to embark on similar research in the future. Our findings highlight the predominance of mental health studies, discuss the opportunities of using standardized sensing approaches and machine-learning advancements, and present the trends of smartphone sensing in healthcare over the years.

Automatic Recognition and Analysis of Balance Activity in Community-Dwelling Older Adults: Algorithm Validation

BACKGROUND

Clinical mobility and balance assessments identify older adults who have a high risk of falls in clinics. In the past two decades, sensors have been a popular supplement to mobility and balance assessment to provide quantitative information and a cost-effective solution in the community environment. Nonetheless, the current sensor-based balance assessment relies on manual observation or motion-specific features to identify motions of research interest.

OBJECTIVE

The objective of this study was to develop an automatic motion data analytics framework using signal data collected from an inertial sensor for balance activity analysis in community-dwelling older adults.

METHODS

In total, 59 community-dwelling older adults (19 males and 40 females; mean age = 81.86 years, SD 6.95 years) were recruited in this study. Data were collected using a body-worn inertial measurement unit (including an accelerometer and a gyroscope) at the L4 vertebra of each individual. After data preprocessing and motion detection via a convolutional long short-term memory (LSTM) neural network, a one-class support vector machine (SVM), linear discriminant analysis (LDA), and k-nearest neighborhood (k-NN) were adopted to classify high-risk individuals.

RESULTS

The framework developed in this study yielded mean accuracies of 87%, 86%, and 89% in detecting sit-to-stand, turning 360°, and stand-to-sit motions, respectively. The balance assessment classification showed accuracies of 90%, 92%, and 86% in classifying abnormal sit-to-stand, turning 360°, and stand-to-sit motions, respectively, using Tinetti Performance Oriented Mobility Assessment-Balance (POMA-B) criteria by the one-class SVM and k-NN.

CONCLUSIONS

The sensor-based approach presented in this study provided a time-effective manner with less human efforts to identify and preprocess the inertial signal and thus enabled an efficient balance assessment tool for medical professionals. In the long run, the approach may offer a flexible solution to relieve the community's burden of continuous health monitoring.

Human Activity Recognition: A Comparative Study to Assess the Contribution Level of Accelerometer, ECG, and PPG Signals

Inertial sensors are widely used in the field of human activity recognition (HAR), since this source of information is the most informative time series among non-visual datasets. HAR researchers are actively exploring other approaches and different sources of signals to improve the performance of HAR systems. In this study, we investigate the impact of combining bio-signals with a dataset acquired from inertial sensors on recognizing human daily activities. To achieve this aim, we used the PPG-DaLiA dataset consisting of 3D-accelerometer (3D-ACC), electrocardiogram (ECG), photoplethysmogram (PPG) signals acquired from 15 individuals while performing daily activities. We extracted hand-crafted time and frequency domain features, then, we applied a correlation-based feature selection approach to reduce the feature-set dimensionality. After introducing early fusion scenarios, we trained and tested random forest models with subject-dependent and subject-independent setups. Our results indicate that combining features extracted from the 3D-ACC signal with the ECG signal improves the classifier's performance F1-scores by 2.72% and 3.00% (from 94.07% to 96.80%, and 83.16% to 86.17%) for subject-dependent and subject-independent approaches, respectively.

Physical Activity Recognition Based on a Parallel Approach for an Ensemble of Machine Learning and Deep Learning Classifiers

Human activity recognition (HAR) by wearable sensor devices embedded in the Internet of things (IOT) can play a significant role in remote health monitoring and emergency notification to provide healthcare of higher standards. The purpose of this study is to investigate a human activity recognition method of accrued decision accuracy and speed of execution to be applicable in healthcare. This method classifies wearable sensor acceleration time series data of human movement using an efficient classifier combination of feature engineering-based and feature learning-based data representation. Leave-one-subject-out cross-validation of the method with data acquired from 44 subjects wearing a single waist-worn accelerometer on a smart textile, and engaged in a variety of 10 activities, yielded an average recognition rate of 90%, performing significantly better than individual classifiers. The method easily accommodates functional and computational parallelization to bring execution time significantly down.

Optimizing Sensor Deployment for Multi-Sensor-Based HAR System with Improved Glowworm Swarm Optimization Algorithm

Human activity recognition (HAR) technology that analyzes and fuses the data acquired from various homogeneous or heterogeneous sensor sources has motivated the development of enormous human-centered applications such as healthcare, fitness, ambient assisted living and rehabilitation. The concurrent use of multiple sensor sources for HAR is a good choice because the plethora of user information provided by the various sensor sources may be useful. However, a multi-sensor system with too many sensors will bring large power consumption and some sensor sources may bring little improvements to the performance. Therefore, the multi-sensor deployment research that can gain a tradeoff among computational complexity and performance is imperative. In this paper, we propose a multi-sensor-based HAR system whose sensor deployment can be optimized by selective ensemble approaches. With respect to optimization of the sensor deployment, an improved binary glowworm swarm optimization (IBGSO) algorithm is proposed and the sensor sources that have a significant effect on the performance of HAR are selected. Furthermore, the ensemble learning system based on optimized sensor deployment is constructed for HAR. Experimental results on two datasets show that the proposed IBGSO-based multi-sensor deployment approach can select a smaller number of sensor sources while achieving better performance than the ensemble of all sensors and other optimization-based selective ensemble approaches.

Enhanced Hand-Oriented Activity Recognition Based on Smartwatch Sensor Data Using LSTMs

The creation of the Internet of Things (IoT), along with the latest developments in wearable technology, has provided new opportunities in human activity recognition (HAR). The modern smartwatch offers the potential for data from sensors to be relayed to novel IoT platforms, which allow the constant tracking and monitoring of human movement and behavior. Recently, traditional activity recognition techniques have done research in advance by choosing machine learning methods such as artificial neural network, decision tree, support vector machine, and naive Bayes. Nonetheless, these conventional machine learning techniques depend inevitably on heuristically handcrafted feature extraction, in which human domain knowledge is normally limited. This work proposes a hybrid deep learning model called CNN-LSTM that employed Long Short-Term Memory (LSTM) networks for activity recognition with the Convolution Neural Network (CNN). The study makes use of HAR involving smartwatches to categorize hand movements. Using the study based on the Wireless Sensor Data Mining (WISDM) public benchmark dataset, the recognition abilities of the deep learning model can be accessed. The accuracy, precision, recall, and F-measure statistics are employed using the evaluation metrics to assess the recognition abilities of LSTM models proposed. The findings indicate that this hybrid deep learning model offers better performance than its rivals, where the achievement of 96.2% accuracy, while the f-measure is 96.3%, is obtained. The results show that the proposed CNN-LSTM can support an improvement of the performance of activity recognition.

Sensor Type, Axis, and Position-Based Fusion and Feature Selection for Multimodal Human Daily Activity Recognition in Wearable Body Sensor Networks

This research addresses the challenge of recognizing human daily activities using surface electromyography (sEMG) and wearable inertial sensors. Effective and efficient recognition in this context has emerged as a cornerstone in robust remote health monitoring systems, among other applications. We propose a novel pipeline that can attain state-of-the-art recognition accuracies on a recent-and-standard dataset—the Human Gait Database (HuGaDB). Using wearable gyroscopes, accelerometers, and electromyography sensors placed on the thigh, shin, and foot, we developed an approach that jointly performs sensor fusion and feature selection. Being done jointly, the proposed pipeline empowers the learned model to benefit from the interaction of features that might have been dropped otherwise. Using statistical and time-based features from heterogeneous signals of the aforementioned sensor types, our approach attains a mean accuracy of 99.8%, which is the highest accuracy on HuGaDB in the literature. This research underlines the potential of incorporating EMG signals especially when fusion and selection are done simultaneously. Meanwhile, it is valid even with simple off-the-shelf feature selection methods such the Sequential Feature Selection family of algorithms. Moreover, through extensive simulations, we show that the left thigh is a key placement for attaining high accuracies. With one inertial sensor on that single placement alone, we were able to achieve a mean accuracy of 98.4%. The presented in-depth comparative analysis shows the influence that every sensor type, position, and placement can have on the attained recognition accuracies—a tool that can facilitate the development of robust systems, customized to specific scenarios and real-life applications.

Application of Modern Multi-Sensor Holter in Diagnosis and Treatment

Modern Holter devices are very trendy tools used in medicine, research, or sport. They monitor a variety of human physiological or pathophysiological signals. Nowadays, Holter devices have been developing very fast. New innovative products come to the market every day. They have become smaller, smarter, cheaper, have ultra-low power consumption, do not limit everyday life, and allow comfortable measurements of humans to be accomplished in a familiar and natural environment, without extreme fear from doctors. People can be informed about their health and 24/7 monitoring can sometimes easily detect specific diseases, which are normally passed during routine ambulance operation. However, there is a problem with the reliability, quality, and quantity of the collected data. In normal life, there may be a loss of signal recording, abnormal growth of artifacts, etc. At this point, there is a need for multiple sensors capturing single variables in parallel by different sensing methods to complement these methods and diminish the level of artifacts. We can also sense multiple different signals that are complementary and give us a coherent picture. In this article, we describe actual interesting multi-sensor principles on the grounds of our own long-year experiences and many experiments.

Development of Rheumatoid Arthritis Classification from Electronic Image Sensor Using Ensemble Method

Rheumatoid arthritis (RA) is an autoimmune illness that impacts the musculoskeletal system by causing chronic, inflammatory, and systemic effects. The disease often becomes progressive and reduces physical function, causes suffering, fatigue, and articular damage. Over a long period of time, RA causes harm to the bone and cartilage of the joints, weakens the joints' muscles and tendons, eventually causing joint destruction. Sensors such as accelerometer, wearable sensors, and thermal infrared camera sensor are widely used to gather data for RA. In this paper, the classification of medical disorders based on RA and orthopaedics datasets using Ensemble methods are discussed. The RA dataset was gathered from the analysis of white blood cell classification using features extracted from the image of lymphocytes acquired from a digital microscope with an electronic image sensor. The orthopaedic dataset is a benchmark dataset for this study, as it posed a similar classification problem with several numerical features. Three ensemble algorithms such as bagging, Adaboost, and random subspace were used in the study. These ensemble classifiers use k-NN (K-nearest neighbours) and Random forest (RF) as the base learners of the ensemble classifiers. The data classification is accessed using holdout and 10-fold cross-validation evaluation methods. The assessment was based on set of performance measures such as precision, recall, F-measure, and receiver operating characteristic (ROC) curve. The performance was also measured based on the comparison of the overall classification accuracy rate between different ensembles classifiers and the base learners. Overall, it was found that for Dataset 1, random subspace classifier with k-NN shows the best results in terms of overall accuracy rate of 97.50% and for Dataset 2, bagging-RF shows the highest overall accuracy rate of 94.84% over different ensemble classifiers. The findings indicate that the efficiency of the base classifiers with ensemble classifier have substantially improved.

Software and Hardware Requirements and Trade-Offs in Operating Systems for Wearables: A Tool to Improve Devices' Performance

Wearable device requirements currently vary from soft to hard real-time constraints. Frequently, hardware improvements are a way to speed-up the global performance of a solution. However, changing some parts or the whole hardware may increase device complexity, raising the costs and leading to development delays of products or research prototypes. This paper focuses on software improvements, presenting a tool designed to create different versions of operating systems (OSs) fitting the specifications of wearable devices projects. Authors have developed a software tool allowing the end-user to craft a new OS in just a few steps. In order to validate the generated OS, an original wearable prototype for mining environments is outlined. Resulting data presented here allows for measuring the actual impact an OS has in different variables of a solution. Finally, the analysis also allows for evaluating the performance impact associated with each hardware part. Results suggest the viability of using the proposed solution when searching for performance improvements on wearables.

A Human Activity Recognition Algorithm Based on Stacking Denoising Autoencoder and LightGBM

Recently, the demand for human activity recognition has become more and more urgent. It is widely used in indoor positioning, medical monitoring, safe driving, etc. Existing activity recognition approaches require either the location information of the sensors or the specific domain knowledge, which are expensive, intrusive, and inconvenient for pervasive implementation. In this paper, a human activity recognition algorithm based on SDAE (Stacking Denoising Autoencoder) and LightGBM (LGB) is proposed. The SDAE is adopted to sanitize the noise in raw sensor data and extract the most effective characteristic expression with unsupervised learning. The LGB reveals the inherent feature dependencies among categories for accurate human activity recognition. Extensive experiments are conducted on four datasets of distinct sensor combinations collected by different devices in three typical application scenarios, which are human moving modes, current static, and dynamic behaviors of users. The experimental results demonstrate that our proposed algorithm achieves an average accuracy of 95.99%, outperforming other comparative algorithms using XGBoost, CNN (Convolutional Neural Network), CNN + Statistical features, or single SDAE.