Accelerometer-Based Human Activity Recognition in Smartphones for Healthcare Services

Classifying tasks performed by electrical line workers using a wrist-worn sensor: A data analytic approach

Electrical line workers (ELWs) experience harsh environments, characterized by long shifts, remote operations, and potentially risky tasks. Wearables present an opportunity for unobtrusive monitoring of productivity and safety. A prerequisite to monitoring is the automated identification of the tasks being performed. Human activity recognition has been widely used for classification for activities of daily living. However, the literature is limited for electrical line maintenance/repair tasks due to task variety and complexity. We investigated how features can be engineered from a single wrist-worn accelerometer for the purpose of classifying ELW tasks. Specifically, three classifiers were investigated across three feature sets (time, frequency, and time-frequency) and two window lengths (4 and 10 seconds) to identify ten common ELW tasks. Based on data from 37 participants in a lab environment, two application scenarios were evaluated: (a) intra-subject, where individualized models were trained and deployed for each worker; and (b) inter-subject, where data was pooled to train a general model that can be deployed for new workers. Accuracies ≥ 93% were achieved for both scenarios, and increased to ≥96% with 10-second windows. Overall and class-specific feature importance were computed, and the impact of those features on the obtained predictions were explained. This work will contribute to the future risk mitigation of ELWs using wearables.

Non-Linear Chaotic Features-Based Human Activity Recognition

Human activity recognition (HAR) has vital applications in human–computer interaction, somatosensory games, and motion monitoring, etc. On the basis of the human motion accelerate sensor data, through a nonlinear analysis of the human motion time series, a novel method for HAR that is based on non-linear chaotic features is proposed in this paper. First, the C-C method and G-P algorithm are used to, respectively, compute the optimal delay time and embedding dimension. Additionally, a Reconstructed Phase Space (RPS) is formed while using time-delay embedding for the human accelerometer motion sensor data. Subsequently, a two-dimensional chaotic feature matrix is constructed, where the chaotic feature is composed of the correlation dimension and largest Lyapunov exponent (LLE) of attractor trajectory in the RPS. Next, the classification algorithms are used in order to classify and recognize the two different activity classes, i.e., basic and transitional activities. The experimental results show that the chaotic feature has a higher accuracy than traditional time and frequency domain features.

Applying Multivariate Segmentation Methods to Human Activity Recognition From Wearable Sensors' Data

BACKGROUND

Time-resolved quantification of physical activity can contribute to both personalized medicine and epidemiological research studies, for example, managing and identifying triggers of asthma exacerbations. A growing number of reportedly accurate machine learning algorithms for human activity recognition (HAR) have been developed using data from wearable devices (eg, smartwatch and smartphone). However, many HAR algorithms depend on fixed-size sampling windows that may poorly adapt to real-world conditions in which activity bouts are of unequal duration. A small sliding window can produce noisy predictions under stable conditions, whereas a large sliding window may miss brief bursts of intense activity.

OBJECTIVE

We aimed to create an HAR framework adapted to variable duration activity bouts by (1) detecting the change points of activity bouts in a multivariate time series and (2) predicting activity for each homogeneous window defined by these change points.

METHODS

We applied standard fixed-width sliding windows (4-6 different sizes) or greedy Gaussian segmentation (GGS) to identify break points in filtered triaxial accelerometer and gyroscope data. After standard feature engineering, we applied an Xgboost model to predict physical activity within each window and then converted windowed predictions to instantaneous predictions to facilitate comparison across segmentation methods. We applied these methods in 2 datasets: the human activity recognition using smartphones (HARuS) dataset where a total of 30 adults performed activities of approximately equal duration (approximately 20 seconds each) while wearing a waist-worn smartphone, and the Biomedical REAl-Time Health Evaluation for Pediatric Asthma (BREATHE) dataset where a total of 14 children performed 6 activities for approximately 10 min each while wearing a smartwatch. To mimic a real-world scenario, we generated artificial unequal activity bout durations in the BREATHE data by randomly subdividing each activity bout into 10 segments and randomly concatenating the 60 activity bouts. Each dataset was divided into ~90% training and ~10% holdout testing.

RESULTS

In the HARuS data, GGS produced the least noisy predictions of 6 physical activities and had the second highest accuracy rate of 91.06% (the highest accuracy rate was 91.79% for the sliding window of size 0.8 second). In the BREATHE data, GGS again produced the least noisy predictions and had the highest accuracy rate of 79.4% of predictions for 6 physical activities.

CONCLUSIONS

In a scenario with variable duration activity bouts, GGS multivariate segmentation produced smart-sized windows with more stable predictions and a higher accuracy rate than traditional fixed-size sliding window approaches. Overall, accuracy was good in both datasets but, as expected, it was slightly lower in the more real-world study using wrist-worn smartwatches in children (BREATHE) than in the more tightly controlled study using waist-worn smartphones in adults (HARuS). We implemented GGS in an offline setting, but it could be adapted for real-time prediction with streaming data.

Promises and Challenges in the Use of Consumer-Grade Devices for Sleep Monitoring

The market for smartphones, smartwatches, and wearable devices is booming. In recent years, individuals and researchers have used these devices as additional tools to monitor and track sleep, physical activity, and behavior. Their use in sleep research and clinical applications could address the difficulties in scaling up studies that rely on polysomnography, the gold-standard. However, the use of commercial devices for large-scale sleep studies is not without challenges. With this in mind, this paper presents an extensive review of sleep monitoring systems and the techniques used in their development. We also discuss their performance in terms of reliability and validity, and consider the needs and expectations of users, whether they are experts, patients, or the general public. Through this review, we highlight a number of challenges with current studies: a lack of standard evaluation methods for consumer-grade devices (e.g., reliability and validity assessment); limitations in the populations studied; consumer expectations of monitoring devices; constraints on the resources of consumer-grade devices (e.g., power consumption).

A Robust and Device-Free System for the Recognition and Classification of Elderly Activities

Human activity recognition, tracking and classification is an essential trend in assisted living systems that can help support elderly people with their daily activities. Traditional activity recognition approaches depend on vision-based or sensor-based techniques. Nowadays, a novel promising technique has obtained more attention, namely device-free human activity recognition that neither requires the target object to wear or carry a device nor install cameras in a perceived area. The device-free technique for activity recognition uses only the signals of common wireless local area network (WLAN) devices available everywhere. In this paper, we present a novel elderly activities recognition system by leveraging the fluctuation of the wireless signals caused by human motion. We present an efficient method to select the correct data from the Channel State Information (CSI) streams that were neglected in previous approaches. We apply a Principle Component Analysis method that exposes the useful information from raw CSI. Thereafter, Forest Decision (FD) is adopted to classify the proposed activities and has gained a high accuracy rate. Extensive experiments have been conducted in an indoor environment to test the feasibility of the proposed system with a total of five volunteer users. The evaluation shows that the proposed system is applicable and robust to electromagnetic noise.