Better physical activity classification using smartphone acceleration sensor

Multi-dimensional task recognition for human-robot teaming: literature review

Human-robot teams collaborating to achieve tasks under various conditions, especially in unstructured, dynamic environments will require robots to adapt autonomously to a human teammate's state. An important element of such adaptation is the robot's ability to infer the human teammate's tasks. Environmentally embedded sensors (e.g., motion capture and cameras) are infeasible in such environments for task recognition, but wearable sensors are a viable task recognition alternative. Human-robot teams will perform a wide variety of composite and atomic tasks, involving multiple activity components (i.e., gross motor, fine-grained motor, tactile, visual, cognitive, speech and auditory) that may occur concurrently. A robot's ability to recognize the human's composite, concurrent tasks is a key requirement for realizing successful teaming. Over a hundred task recognition algorithms across multiple activity components are evaluated based on six criteria: sensitivity, suitability, generalizability, composite factor, concurrency and anomaly awareness. The majority of the reviewed task recognition algorithms are not viable for human-robot teams in unstructured, dynamic environments, as they only detect tasks from a subset of activity components, incorporate non-wearable sensors, and rarely detect composite, concurrent tasks across multiple activity components.

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.

Ensemble of RNN Classifiers for Activity Detection Using a Smartphone and Supporting Nodes

Nowadays, sensor-equipped mobile devices allow us to detect basic daily activities accurately. However, the accuracy of the existing activity recognition methods decreases rapidly if the set of activities is extended and includes training routines, such as squats, jumps, or arm swings. Thus, this paper proposes a model of a personal area network with a smartphone (as a main node) and supporting sensor nodes that deliver additional data to increase activity-recognition accuracy. The introduced personal area sensor network takes advantage of the information from multiple sensor nodes attached to different parts of the human body. In this scheme, nodes process their sensor readings locally with the use of recurrent neural networks (RNNs) to categorize the activities. Then, the main node collects results from supporting sensor nodes and performs a final activity recognition run based on a weighted voting procedure. In order to save energy and extend the network's lifetime, sensor nodes report their local results only for specific types of recognized activity. The presented method was evaluated during experiments with sensor nodes attached to the waist, chest, leg, and arm. The results obtained for a set of eight activities show that the proposed approach achieves higher recognition accuracy when compared with the existing methods. Based on the experimental results, the optimal configuration of the sensor nodes was determined to maximize the activity-recognition accuracy and reduce the number of transmissions from supporting sensor nodes.

A Comparative Study on the Influence of Undersampling and Oversampling Techniques for the Classification of Physical Activities Using an Imbalanced Accelerometer Dataset

Accelerometer data collected from wearable devices have recently been used to monitor physical activities (PAs) in daily life. While the intensity of PAs can be distinguished with a cut-off approach, it is important to discriminate different behaviors with similar accelerometry patterns to estimate energy expenditure. We aim to overcome the data imbalance problem that negatively affects machine learning-based PA classification by extracting well-defined features and applying undersampling and oversampling methods. We extracted various temporal, spectral, and nonlinear features from wrist-, hip-, and ankle-worn accelerometer data. Then, the influences of undersampilng and oversampling were compared using various ML and DL approaches. Among various ML and DL models, ensemble methods including random forest (RF) and adaptive boosting (AdaBoost) exhibited great performance in differentiating sedentary behavior (driving) and three walking types (walking on level ground, ascending stairs, and descending stairs) even in a cross-subject paradigm. The undersampling approach, which has a low computational cost, exhibited classification results unbiased to the majority class. In addition, we found that RF could automatically select relevant features for PA classification depending on the sensor location by examining the importance of each node in multiple decision trees (DTs). This study proposes that ensemble learning using well-defined feature sets combined with the undersampling approach is robust for imbalanced datasets in PA classification. This approach will be useful for PA classification in the free-living situation, where data imbalance problems between classes are common.

A systematic review of smartphone-based human activity recognition methods for health research

Smartphones are now nearly ubiquitous; their numerous built-in sensors enable continuous measurement of activities of daily living, making them especially well-suited for health research. Researchers have proposed various human activity recognition (HAR) systems aimed at translating measurements from smartphones into various types of physical activity. In this review, we summarized the existing approaches to smartphone-based HAR. For this purpose, we systematically searched Scopus, PubMed, and Web of Science for peer-reviewed articles published up to December 2020 on the use of smartphones for HAR. We extracted information on smartphone body location, sensors, and physical activity types studied and the data transformation techniques and classification schemes used for activity recognition. Consequently, we identified 108 articles and described the various approaches used for data acquisition, data preprocessing, feature extraction, and activity classification, identifying the most common practices, and their alternatives. We conclude that smartphones are well-suited for HAR research in the health sciences. For population-level impact, future studies should focus on improving the quality of collected data, address missing data, incorporate more diverse participants and activities, relax requirements about phone placement, provide more complete documentation on study participants, and share the source code of the implemented methods and algorithms.

Smartphone-Based Human Sitting Behaviors Recognition Using Inertial Sensor

At present, people spend most of their time in passive rather than active mode. Sitting with computers for a long time may lead to unhealthy conditions like shoulder pain, numbness, headache, etc. To overcome this problem, human posture should be changed for particular intervals of time. This paper deals with using an inertial sensor built in the smartphone and can be used to overcome the unhealthy human sitting behaviors (HSBs) of the office worker. To monitor, six volunteers are considered within the age band of 26 ± 3 years, out of which four were male and two were female. Here, the inertial sensor is attached to the rear upper trunk of the body, and a dataset is generated for five different activities performed by the subjects while sitting in the chair in the office. Correlation-based feature selection (CFS) technique and particle swarm optimization (PSO) methods are jointly used to select feature vectors. The optimized features are fed to machine learning supervised classifiers such as naive Bayes, SVM, and KNN for recognition. Finally, the SVM classifier achieved 99.90% overall accuracy for different human sitting behaviors using an accelerometer, gyroscope, and magnetometer sensors.

Activity recognition from smartphone data using weighted learning methods

Mobile phone based activity recognition uses data obtained from embedded sensors to infer user’s physical activities. The traditional approach for activity recognition employs machine learning algorithms to learn from collected labeled data and induce a model. To enhance the accuracy and hence to improve the overall efficiency of the system, the good classifiers can be combined together. Fusion can be done at the feature level and also at the decision level. In this work, we propose a new hybrid classification model Weighted SVM-KNN to perform automatic recognition of activities that combines a Weighted Support Vector Machines (WSVM) to learn a model with a Weighted K-Nearest Neighbors (WKNN), to classify and identify the ongoing activity. The sensory inputs to the classifier are reduced with the Linear Discriminant Analysis (LDA). We demonstrate how to train the hybrid approach in this setting, introduce an adaptive regularization parameter for WSVM approach, and illustrate how our method outperforms the state-of-the-art on a large benchmark datasets.

Development of a Wearable Camera and AI Algorithm for Medication Behavior Recognition

As many as 40% to 50% of patients do not adhere to long-term medications for managing chronic conditions, such as diabetes or hypertension. Limited opportunity for medication monitoring is a major problem from the perspective of health professionals. The availability of prompt medication error reports can enable health professionals to provide immediate interventions for patients. Furthermore, it can enable clinical researchers to modify experiments easily and predict health levels based on medication compliance. This study proposes a method in which videos of patients taking medications are recorded using a camera image sensor integrated into a wearable device. The collected data are used as a training dataset based on applying the latest convolutional neural network (CNN) technique. As for an artificial intelligence (AI) algorithm to analyze the medication behavior, we constructed an object detection model (Model 1) using the faster region-based CNN technique and a second model that uses the combined feature values to perform action recognition (Model 2). Moreover, 50,000 image data were collected from 89 participants, and labeling was performed on different data categories to train the algorithm. The experimental combination of the object detection model (Model 1) and action recognition model (Model 2) was newly developed, and the accuracy was 92.7%, which is significantly high for medication behavior recognition. This study is expected to enable rapid intervention for providers seeking to treat patients through rapid reporting of drug errors.

w-HAR: An Activity Recognition Dataset and Framework Using Low-Power Wearable Devices

Human activity recognition (HAR) is growing in popularity due to its wide-ranging applications in patient rehabilitation and movement disorders. HAR approaches typically start with collecting sensor data for the activities under consideration and then develop algorithms using the dataset. As such, the success of algorithms for HAR depends on the availability and quality of datasets. Most of the existing work on HAR uses data from inertial sensors on wearable devices or smartphones to design HAR algorithms. However, inertial sensors exhibit high noise that makes it difficult to segment the data and classify the activities. Furthermore, existing approaches typically do not make their data available publicly, which makes it difficult or impossible to obtain comparisons of HAR approaches. To address these issues, we present wearable HAR (w-HAR) which contains labeled data of seven activities from 22 users. Our dataset’s unique aspect is the integration of data from inertial and wearable stretch sensors, thus providing two modalities of activity information. The wearable stretch sensor data allows us to create variable-length segment data and ensure that each segment contains a single activity. We also provide a HAR framework to use w-HAR to classify the activities. To this end, we first perform a design space exploration to choose a neural network architecture for activity classification. Then, we use two online learning algorithms to adapt the classifier to users whose data are not included at design time. Experiments on the w-HAR dataset show that our framework achieves 95% accuracy while the online learning algorithms improve the accuracy by as much as 40%.

Augmented Movelet Method for Activity Classification Using Smartphone Gyroscope and Accelerometer Data

Physical activity, such as walking and ascending stairs, is commonly used in biomedical settings as an outcome or covariate. Researchers have traditionally relied on surveys to quantify activity levels of subjects in both research and clinical settings, but surveys are subjective in nature and have known limitations, such as recall bias. Smartphones provide an opportunity for unobtrusive objective measurement of physical activity in naturalistic settings, but their data tends to be noisy and needs to be analyzed with care. We explored the potential of smartphone accelerometer and gyroscope data to distinguish between walking, sitting, standing, ascending stairs, and descending stairs. We conducted a study in which four participants followed a study protocol and performed a sequence of activities with one phone in their front pocket and another phone in their back pocket. The subjects were filmed throughout, and the obtained footage was annotated to establish moment-by-moment ground truth activity. We introduce a modified version of the so-called movelet method to classify activity type and to quantify the uncertainty present in that classification. Our results demonstrate the promise of smartphones for activity recognition in naturalistic settings, but they also highlight challenges in this field of research.

From Bits of Data to Bits of Knowledge-An On-Board Classification Framework for Wearable Sensing Systems

Wearable systems constitute a promising solution to the emerging challenges of healthcare provision, feeding machine learning frameworks with necessary data. In practice, however, raw data collection is expensive in terms of energy, and therefore imposes a significant maintenance burden to the user, which in turn results in poor user experience, as well as significant data loss due to improper battery maintenance. In this paper, we propose a framework for on-board activity classification targeting severely energy-constrained wearable systems. The proposed framework leverages embedded classifiers to activate power-hungry sensing elements only when they are useful, and to distil the raw data into knowledge that is eventually transmitted over the air. We implement the proposed framework on a prototype wearable system and demonstrate that it can decrease the energy requirements by one order of magnitude, yielding high classification accuracy that is reduced by approximately 5%, as compared to a cloud-based reference system.

Passive Sensing of Health Outcomes Through Smartphones: Systematic Review of Current Solutions and Possible Limitations

BACKGROUND

Technological advancements, together with the decrease in both price and size of a large variety of sensors, has expanded the role and capabilities of regular mobile phones, turning them into powerful yet ubiquitous monitoring systems. At present, smartphones have the potential to continuously collect information about the users, monitor their activities and behaviors in real time, and provide them with feedback and recommendations.

OBJECTIVE

This systematic review aimed to identify recent scientific studies that explored the passive use of smartphones for generating health- and well-being-related outcomes. In addition, it explores users' engagement and possible challenges in using such self-monitoring systems.

METHODS

A systematic review was conducted, following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, to identify recent publications that explore the use of smartphones as ubiquitous health monitoring systems. We ran reproducible search queries on PubMed, IEEE Xplore, ACM Digital Library, and Scopus online databases and aimed to find answers to the following questions: (1) What is the study focus of the selected papers? (2) What smartphone sensing technologies and data are used to gather health-related input? (3) How are the developed systems validated? and (4) What are the limitations and challenges when using such sensing systems?

RESULTS

Our bibliographic research returned 7404 unique publications. Of these, 118 met the predefined inclusion criteria, which considered publication dates from 2014 onward, English language, and relevance for the topic of this review. The selected papers highlight that smartphones are already being used in multiple health-related scenarios. Of those, physical activity (29.6%; 35/118) and mental health (27.9; 33/118) are 2 of the most studied applications. Accelerometers (57.7%; 67/118) and global positioning systems (GPS; 40.6%; 48/118) are 2 of the most used sensors in smartphones for collecting data from which the health status or well-being of its users can be inferred.

CONCLUSIONS

One relevant outcome of this systematic review is that although smartphones present many advantages for the passive monitoring of users' health and well-being, there is a lack of correlation between smartphone-generated outcomes and clinical knowledge. Moreover, user engagement and motivation are not always modeled as prerequisites, which directly affects user adherence and full validation of such systems.

Physical Activity Classification for Elderly People in Free-Living Conditions

Physical activity is strongly linked with mental and physical health in the elderly population and accurate monitoring of activities of daily living (ADLs) can help improve quality of life and well-being. This study presents and validates an inertial sensors-based physical activity classification system developed with older adults as the target population. The dataset was collected in free-living conditions without placing constraints on the way and order of performing ADLs. Four sensor locations (chest, lower back, wrist, and thigh) were explored to obtain the optimal number and combination of sensors by finding the best tradeoff between the system's performance and wearability. Several feature selection techniques were implemented on the feature set obtained from acceleration and angular velocity signals to classify four major ADLs (sitting, standing, walking, and lying). A support vector machine was used for the classification of the ADLs. The findings show the potential of different solutions (single sensor or multisensor) to correctly classify the ADLs of older people in free-living conditions. Considering a minimal set-up of a single sensor, the sensor worn at the L5 achieved the best performance. A two-sensor solution (L5 + thigh) achieved a better performance with respect to a single-sensor solution. By contrast, considering more than two sensors did not provide further improvements. Finally, we evaluated the computational cost of different solutions and it was shown that a feature selection step can reduce the computational cost of the system and increase the system performance in most cases. This can be helpful for real-time applications.

Performance Evaluation of State of the Art Systems for Physical Activity Classification of Older Subjects Using Inertial Sensors in a Real Life Scenario: A Benchmark Study

The popularity of using wearable inertial sensors for physical activity classification has dramatically increased in the last decade due to their versatility, low form factor, and low power requirements. Consequently, various systems have been developed to automatically classify daily life activities. However, the scope and implementation of such systems is limited to laboratory-based investigations. Furthermore, these systems are not directly comparable, due to the large diversity in their design (e.g., number of sensors, placement of sensors, data collection environments, data processing techniques, features set, classifiers, cross-validation methods). Hence, the aim of this study is to propose a fair and unbiased benchmark for the field-based validation of three existing systems, highlighting the gap between laboratory and real-life conditions. For this purpose, three representative state-of-the-art systems are chosen and implemented to classify the physical activities of twenty older subjects (76.4 ± 5.6 years). The performance in classifying four basic activities of daily life (sitting, standing, walking, and lying) is analyzed in controlled and free living conditions. To observe the performance of laboratory-based systems in field-based conditions, we trained the activity classification systems using data recorded in a laboratory environment and tested them in real-life conditions in the field. The findings show that the performance of all systems trained with data in the laboratory setting highly deteriorates when tested in real-life conditions, thus highlighting the need to train and test the classification systems in the real-life setting. Moreover, we tested the sensitivity of chosen systems to window size (from 1 s to 10 s) suggesting that overall accuracy decreases with increasing window size. Finally, to evaluate the impact of the number of sensors on the performance, chosen systems are modified considering only the sensing unit worn at the lower back. The results, similarly to the multi-sensor setup, indicate substantial degradation of the performance when laboratory-trained systems are tested in the real-life setting. This degradation is higher than in the multi-sensor setup. Still, the performance provided by the single-sensor approach, when trained and tested with real data, can be acceptable (with an accuracy above 80%).

An Automatic User-Adapted Physical Activity Classification Method Using Smartphones

In recent years, an increasing number of people have become concerned about their health. Most chronic diseases are related to lifestyle, and daily activity records can be used as an important indicator of health. Specifically, using advanced technology to automatically monitor actual activities can effectively prevent and manage chronic diseases. The data used in this paper were obtained from acceleration sensors and gyroscopes integrated in smartphones. We designed an efficient Adaboost-Stump running on a smartphone to classify five common activities: cycling, running, sitting, standing, and walking and achieved a satisfactory classification accuracy of 98%. We designed an online learning method, and the classification model requires continuous training with actual data. The parameters in the model then become increasingly fitted to the specific user, which allows the classification accuracy to reach 95% under different use environments. In addition, this paper also utilized the OpenCL framework to design the program in parallel. This process can enhance the computing efficiency approximately ninefold.

Mobile Phone and Web 2.0 Technologies for Weight Management: A Systematic Scoping Review

BACKGROUND

Widespread diffusion of mobile phone and Web 2.0 technologies make them potentially useful tools for promoting health and tackling public health issues, such as the increasing prevalence of overweight and obesity. Research in this domain is growing rapidly but, to date, no review has comprehensively and systematically documented how mobile and Web 2.0 technologies are being deployed and evaluated in relation to weight management.

OBJECTIVE

To provide an up-to-date, comprehensive map of the literature discussing the use of mobile phone and Web 2.0 apps for influencing behaviors related to weight management (ie, diet, physical activity [PA], weight control, etc).

METHODS

A systematic scoping review of the literature was conducted based on a published protocol (registered at

PROSPERO

CRD42014010323). Using a comprehensive search strategy, we searched 16 multidisciplinary electronic databases for original research documents published in English between 2004 and 2014. We used duplicate study selection and data extraction. Using an inductively developed charting tool, selected articles were thematically categorized.

RESULTS

We identified 457 articles, mostly published between 2013 and 2014 in 157 different journals and 89 conference proceedings. Articles were categorized around two overarching themes, which described the use of technologies for either (1) promoting behavior change (309/457, 67.6%) or (2) measuring behavior (103/457, 22.5%). The remaining articles were overviews of apps and social media content (33/457, 7.2%) or covered a combination of these three themes (12/457, 2.6%). Within the two main overarching themes, we categorized articles as representing three phases of research development: (1) design and development, (2) feasibility studies, and (3) evaluations. Overall, articles mostly reported on evaluations of technologies for behavior change (211/457, 46.2%).

CONCLUSIONS

There is an extensive body of research on mobile phone and Web 2.0 technologies for weight management. Research has reported on (1) the development, feasibility, and efficacy of persuasive mobile technologies used in interventions for behavior change (PA and diet) and (2) the design, feasibility, and accuracy of mobile phone apps for behavioral assessment. Further research has focused exclusively on analyses of the content and quality of available apps. Limited evidence exists on the use of social media for behavior change, but a segment of studies deal with content analyses of social media. Future research should analyze mobile phone and Web 2.0 technologies together by combining the evaluation of content and design aspects with usability, feasibility, and efficacy/effectiveness for behavior change, or accuracy/validity for behavior assessment, in order to understand which technological components and features are likely to result in effective interventions.

Perioperative Smartphone Apps and Devices for Patient-Centered Care

Smartphones have grown in ubiquity and computing power, and they play an ever-increasing role in patient-centered health care. The "medicalized smartphone" not only enables web-based access to patient health resources, but also can run patient-oriented software applications and be connected to health-related peripheral devices. A variety of patient-oriented smartphone apps and devices are available for use to facilitate patient-centered care throughout the continuum of perioperative care. Ongoing advances in smartphone technology and health care apps and devices should expand their utility for enhancing patient-centered care in the future.

A Wearable Context-Aware ECG Monitoring System Integrated with Built-in Kinematic Sensors of the Smartphone

Continuously monitoring the ECG signals over hours combined with activity status is very important for preventing cardiovascular diseases. A traditional ECG holter is often inconvenient to carry because it has many electrodes attached to the chest and because it is heavy. This work proposes a wearable, low power context-aware ECG monitoring system integrated built-in kinetic sensors of the smartphone with a self-designed ECG sensor. The wearable ECG sensor is comprised of a fully integrated analog front-end (AFE), a commercial micro control unit (MCU), a secure digital (SD) card, and a Bluetooth module. The whole sensor is very small with a size of only 58 × 50 × 10 mm for wearable monitoring application due to the AFE design, and the total power dissipation in a full round of ECG acquisition is only 12.5 mW. With the help of built-in kinetic sensors of the smartphone, the proposed system can compute and recognize user’s physical activity, and thus provide context-aware information for the continuous ECG monitoring. The experimental results demonstrated the performance of proposed system in improving diagnosis accuracy for arrhythmias and identifying the most common abnormal ECG patterns in different activities. In conclusion, we provide a wearable, accurate and energy-efficient system for long-term and context-aware ECG monitoring without any extra cost on kinetic sensor design but with the help of the widespread smartphone.

An adaptive Hidden Markov model for activity recognition based on a wearable multi-sensor device

Human activity recognition is important in the study of personal health, wellness and lifestyle. In order to acquire human activity information from the personal space, many wearable multi-sensor devices have been developed. In this paper, a novel technique for automatic activity recognition based on multi-sensor data is presented. In order to utilize these data efficiently and overcome the big data problem, an offline adaptive-Hidden Markov Model (HMM) is proposed. A sensor selection scheme is implemented based on an improved Viterbi algorithm. A new method is proposed that incorporates personal experience into the HMM model as a priori information. Experiments are conducted using a personal wearable computer eButton consisting of multiple sensors. Our comparative study with the standard HMM and other alternative methods in processing the eButton data have shown that our method is more robust and efficient, providing a useful tool to evaluate human activity and lifestyle.

Identifying typical physical activity on smartphone with varying positions and orientations

Background

Traditional activity recognition solutions are not widely applicable due to a high cost and inconvenience to use with numerous sensors. This paper aims to automatically recognize physical activity with the help of the built-in sensors of the widespread smartphone without any limitation of firm attachment to the human body.

Methods

By introducing a method to judge whether the phone is in a pocket, we investigated the data collected from six positions of seven subjects, chose five signals that are insensitive to orientation for activity classification. Decision trees (J48), Naive Bayes and Sequential minimal optimization (SMO) were employed to recognize five activities: static, walking, running, walking upstairs and walking downstairs.

Results

The experimental results based on 8,097 activity data demonstrated that the J48 classifier produced the best performance with an average recognition accuracy of 89.6% during the three classifiers, and thus would serve as the optimal online classifier.

Conclusions

The utilization of the built-in sensors of the smartphone to recognize typical physical activities without any limitation of firm attachment is feasible.

Auto detection and segmentation of physical activities during a Timed-Up-and-Go (TUG) task in healthy older adults using multiple inertial sensors

Background

Recently, much attention has been given to the use of inertial sensors for remote monitoring of individuals with limited mobility. However, the focus has been mostly on the detection of symptoms, not specific activities. The objective of the present study was to develop an automated recognition and segmentation algorithm based on inertial sensor data to identify common gross motor patterns during activity of daily living.

Method

A modified Time-Up-And-Go (TUG) task was used since it is comprised of four common daily living activities; Standing, Walking, Turning, and Sitting, all performed in a continuous fashion resulting in six different segments during the task. Sixteen healthy older adults performed two trials of a 5 and 10 meter TUG task. They were outfitted with 17 inertial motion sensors covering each body segment. Data from the 10 meter TUG were used to identify pertinent sensors on the trunk, head, hip, knee, and thigh that provided suitable data for detecting and segmenting activities associated with the TUG. Raw data from sensors were detrended to remove sensor drift, normalized, and band pass filtered with optimal frequencies to reveal kinematic peaks that corresponded to different activities. Segmentation was accomplished by identifying the time stamps of the first minimum or maximum to the right and the left of these peaks. Segmentation time stamps were compared to results from two examiners visually segmenting the activities of the TUG.

Results

We were able to detect these activities in a TUG with 100% sensitivity and specificity (n = 192) during the 10 meter TUG. The rate of success was subsequently confirmed in the 5 meter TUG (n = 192) without altering the parameters of the algorithm. When applying the segmentation algorithms to the 10 meter TUG, we were able to parse 100% of the transition points (n = 224) between different segments that were as reliable and less variable than visual segmentation performed by two independent examiners.

Conclusions

The present study lays the foundation for the development of a comprehensive algorithm to detect and segment naturalistic activities using inertial sensors, in hope of evaluating automatically motor performance within the detected tasks.