BACKGROUND: Physical activity is emerging as an outcome measure. Accelerometers have become an important tool in monitoring physical behavior, and newer analytical approaches of recognition methods increase the degree of details. Many studies have achieved high performance in the classification of physical behaviors through the use of multiple wearable sensors; however, multiple wearables can be impractical and lower compliance.
OBJECTIVE: The aim of this study was to develop and validate an algorithm for classifying several daily physical behaviors using a single thigh-mounted accelerometer and a supervised machine-learning scheme.
METHODS: We collected training data by adding the behavior classes-running, cycling, stair climbing, wheelchair ambulation, and vehicle driving-to an existing algorithm with the classes of sitting, lying, standing, walking, and transitioning. After combining the training data, we used a random forest learning scheme for model development. We validated the algorithm through a simulated free-living procedure using chest-mounted cameras for establishing the ground truth. Furthermore, we adjusted our algorithm and compared the performance with an existing algorithm based on vector thresholds.
RESULTS: We developed an algorithm to classify 11 physical behaviors relevant for rehabilitation. In the simulated free-living validation, the performance of the algorithm decreased to 57% as an average for the 11 classes (F-measure). After merging classes into sedentary behavior, standing, walking, running, and cycling, the result revealed high performance in comparison to both the ground truth and the existing algorithm.
CONCLUSIONS: Using a single thigh-mounted accelerometer, we obtained high classification levels within specific behaviors. The behaviors classified with high levels of performance mostly occur in populations with higher levels of functioning. Further development should aim at describing behaviors within populations with lower levels of functioning.

暂无摘要。

Hand-hygiene is a critical component for safe food handling. In this paper, we apply an iterative engineering process to design a hand-hygiene action detection system to improve food-handling safety. We demonstrate the feasibility of a baseline RGB-only convolutional neural network (CNN) in the restricted case of a single scenario; however, since this baseline system performs poorly across scenarios, we also demonstrate the application of two methods to explore potential reasons for its poor performance. This leads to the development of our hierarchical system that incorporates a variety of modalities (RGB, optical flow, hand masks, and human skeleton joints) for recognizing subsets of hand-hygiene actions. Using hand-washing video recorded from several locations in a commercial kitchen, we demonstrate the effectiveness of our system for detecting hand hygiene actions in untrimmed videos. In addition, we discuss recommendations for designing a computer vision system for a real application.

Machine-learning (ML) approaches have been repeatedly coupled with raw accelerometry to classify physical activity classes, but the features required to optimize their predictive performance are still unknown. Our aim was to identify appropriate combination of feature subsets and prediction algorithms for activity class prediction from hip-based raw acceleration data.
The hip-based raw acceleration data collected from 27 participants was split into training (70 %) and validation (30 %) subsets. A total of 206 time- (TD) and frequencydomain (FD) features were extracted from 6-second non-overlapping windows of the signal. Feature selection was done using seven filter-based, two wrapper-based, and one embedded algorithm, and classification was performed with artificial neural network (ANN), support vector machine (SVM), and random forest (RF). For every combination between the feature selection method and the classifiers, the most appropriate feature subsets were found and used for model training within the training set. These models were then validated with the left-out validation set.
The appropriate number of features for the ANN, SVM, and RF ranged from 20 to 45. Overall, the accuracy of all the three classifiers was higher when trained with feature subsets generated using filter-based methods compared with when they were trained with wrapper-based methods (range: 78.1 %-88 % vs. 66 %-83.5 %). TD features that reflect how signals vary around the mean, how they differ with one another, and how much and how often they change were more frequently selected via the feature selection methods.
A subset of TD features from raw accelerometry could be sufficient for ML-based activity classification if properly selected from different axes.

Activity Recognition (AR) is an active research topic focused on detecting human actions and behaviours in smart environments. In this work, we present the on-line activity recognition platform DOLARS (Distributed On-line Activity Recognition System) where data from heterogeneous sensors are evaluated in real time, including binary, wearable and location sensors. Different descriptors and metrics from the heterogeneous sensor data are integrated in a common feature vector whose extraction is developed by a sliding window approach under real-time conditions. DOLARS provides a distributed architecture where: (i) stages for processing data in AR are deployed in distributed nodes, (ii) temporal cache modules compute metrics which aggregate sensor data for computing feature vectors in an efficient way; (iii) publish-subscribe models are integrated both to spread data from sensors and orchestrate the nodes (communication and replication) for computing AR and (iv) machine learning algorithms are used to classify and recognize the activities. A successful case study of daily activities recognition developed in the Smart Lab of The University of Almería (UAL) is presented in this paper. Results present an encouraging performance in recognition of sequences of activities and show the need for distributed architectures to achieve real time recognition.

UNASSIGNED: Wearable activity trackers are regarded as a new opportunity to deliver health promotion interventions. Indeed, while the prediction of active behaviors is currently primarily relying on the processing of accelerometer sensor data, the emergence of smart clothes with multi-sensing capacities is offering new possibilities. Algorithms able to process data from a variety of smart devices and classify daily life activities could therefore be of particular importance to achieve a more accurate evaluation of physical behaviors. This study aims to (1) develop an activity recognition algorithm based on the processing of plantar pressure information provided by a smart-shoe prototype and (2) to determine the optimal hardware and software configurations.
UNASSIGNED: Seventeen subjects wore a pair of smart-shoe prototypes composed of plantar pressure measurement insoles, and they performed the following nine activities: sitting, standing, walking on a flat surface, walking upstairs, walking downstairs, walking up a slope, running, cycling, and completing office work. The insole featured seven pressure sensors. For each activity, at least four minutes of plantar pressure data were collected. The plantar pressure data were cut in overlapping windows of different lengths and 167 features were extracted for each window. Data were split into training and test samples using a subject-wise assignment method. A random forest model was trained to recognize activity. The resulting activity recognition algorithms were evaluated on the test sample. A multi hold-out procedure allowed repeating the operation with 5 different assignments. The analytic conditions were modulated to test (1) different window lengths (1-60 seconds), (2) some selected sensor configurations and (3) different numbers of data features.
UNASSIGNED: A window length of 20 s was found to be optimum and therefore used for the rest of the analysis. Using all the sensors and all 167 features, the smart shoes predicted the activities with an average success of 89%. \"Running\" demonstrated the highest sensitivity (100%). \"Walking up a slope\" was linked with the lowest performance (63%), with the majority of the false negatives being \"walking on a flat surface\" and \"walking upstairs.\" Some 2- and 3-sensor configurations were linked with an average success rate of 87%. Reducing the number of features down to 20 does not alter significantly the performance of the algorithm.
UNASSIGNED: High-performance human behavior recognition using plantar pressure data only is possible. In the future, smart-shoe devices could contribute to the evaluation of daily physical activities. Minimalist configurations integrating only a small number of sensors and computing a reduced number of selected features could maintain a satisfying performance. Future experiments must include a more heterogeneous population.

BACKGROUND: Orientation deficits are among the most devastating consequences of early dementia. Digital navigation devices could overcome these deficits if adaptable to the user\'s needs (ie, provide situation-aware, proactive navigation assistance). To fulfill this task, systems need to automatically detect spatial disorientation from sensors in real time. Ideally, this would require field studies consisting of real-world navigation. However, such field studies can be challenging and are not guaranteed to cover sufficient instances of disorientation due to the large variability of real-world settings and a lack of control over the environment.
OBJECTIVE: Extending a foregoing field study, we aim to evaluate the feasibility of using a sophisticated virtual reality (VR) setup, which allows a more controlled observation of disorientation states and accompanying behavioral and physiological parameters in cognitively healthy older people and people with dementia.
METHODS: In this feasibility study, we described the experimental design and pilot outcomes of an ongoing study aimed at investigating the effect of disorientation on gait and selected physiological features in a virtual laboratory. We transferred a real-world navigation task to a treadmill-based virtual system for gait analysis. Disorientation was induced by deliberately manipulating landmarks in the VR projection. Associated responses in motion behavior and physiological parameters were recorded by sensors. Primary outcomes were variations in motion and physiological parameters, frequency of disorientation, and questionnaire-derived usability estimates (immersion and perceived control of the gait system) for our population of interest. At this time, the included participants were 9 cognitively healthy older participants [5/9 women, 4/9 men; mean age 70 years, SD 4.40; Mini-Mental State Examination (MMSE) mean 29, SD 0.70) and 4 participants with dementia (2/4 women, 2/4 men; mean age 78 years, SD 2.30 years; MMSE mean 20.50, SD 7.54). Recruitment is ongoing, with the aim of including 30 cognitively healthy older participants and 20 participants with dementia.
RESULTS: All 13 participants completed the experiment. Patients\' route was adapted by shortening it relative to the original route. Average instances of disorientation were 21.40, 36.50, and 37.50 for the cognitively healthy older control, cognitively healthy older experimental participants, and participants with dementia, respectively. Questionnaire outcomes indicated that participants experienced adequate usability and immersion; 4.30 for presence, 3.73 for involvement, and 3.85 for realism of 7 possible points, indicating a good overall ability to cope with the experiment. Variations were also observed in motion and physiological parameters during instances of disorientation.
CONCLUSIONS: This study presents the first feasibility outcomes of a study investigating the viability of using a sophisticated VR setup, based on an earlier real-world navigation study, to study spatial disorientation among cognitively healthy older people and people with dementia. Preliminary outcomes give confidence to the notion that our setup can be used to assess motion and physiological markers of disorientation, even in people with cognitive decline.
BACKGROUND: ClinicalTrials.gov; https://clinicaltrials.gov/ct2/show/NCT04134806.

Activity and emotion recognition based on physiological signal processing in health care applications is a relevant research field, with promising future and relevant applications, such as health at work or preventive care. This paper carries out a deep analysis of features proposed to extract information from the electrocardiogram, thoracic electrical bioimpedance, and electrodermal activity signals. The activities analyzed are: neutral, emotional, mental and physical. A total number of 533 features are tested for activity recognition, performing a comprehensive study taking into consideration the prediction accuracy, feature calculation, window length, and type of classifier. Feature selection to know the most relevant features from the complete set is implemented using a genetic algorithm, with a different number of features. This study has allowed us to determine the best number of features to obtain a good error probability avoiding over-fitting, and the best subset of features among those proposed in the literature. The lowest error probability that is obtained is 22.2%, with 40 features, a least squares error classifier, and 40 seconds window length.

Quality of life (QoL) indicators are now being adopted as clinical outcomes in clinical trials on cancer treatments. Technology-free daily monitoring of patients is complicated, time-consuming and expensive due to the need for vast amounts of resources and personnel. The alternative method of using the patients\' own phones could reduce the burden of continuous monitoring of cancer patients in clinical trials. This paper proposes monitoring the patients\' QoL by gathering data from their own phones. We considered that the continuous multiparametric acquisition of movement, location, phone calls, conversations and data use could be employed to simultaneously monitor their physical, psychological, social and environmental aspects. An open access phone app was developed (Human Dynamics Reporting Service (HDRS)) to implement this approach. We here propose a novel mapping between the standardized QoL items for these patients, the European Organization for the Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ-C30) and define HDRS monitoring indicators. A pilot study with university volunteers verified the plausibility of detecting human activity indicators directly related to QoL.

Wearable accelerometers have greatly improved measurement of physical activity, and the increasing popularity of smartwatches with inherent acceleration data collection suggest their potential use in the physical activity research domain; however, their use needs to be validated.
This study aimed to assess the validity of accelerometer data collected from a Samsung Gear S smartwatch (SGS) compared with an ActiGraph GT3X+ (GT3X+) activity monitor. The study aims were to (1) assess SGS validity using a mechanical shaker; (2) assess SGS validity using a treadmill running test; and (3) compare individual activity recognition, location of major body movement detection, activity intensity detection, locomotion recognition, and metabolic equivalent scores (METs) estimation between the SGS and GT3X+.
To validate and compare the SGS accelerometer data with GT3X+ data, we collected data simultaneously from both devices during highly controlled, mechanically simulated, and less-controlled natural wear conditions. First, SGS and GT3X+ data were simultaneously collected from a mechanical shaker and an individual ambulating on a treadmill. Pearson correlation was calculated for mechanical shaker and treadmill experiments. Finally, SGS and GT3X+ data were simultaneously collected during 15 common daily activities performed by 40 participants (n=12 males, mean age 55.15 [SD 17.8] years). A total of 15 frequency- and time-domain features were extracted from SGS and GT3X+ data. We used these features for training machine learning models on 6 tasks: (1) individual activity recognition, (2) activity intensity detection, (3) locomotion recognition, (4) sedentary activity detection, (5) major body movement location detection, and (6) METs estimation. The classification models included random forest, support vector machines, neural networks, and decision trees. The results were compared between devices. We evaluated the effect of different feature extraction window lengths on model accuracy as defined by the percentage of correct classifications. In addition to these classification tasks, we also used the extracted features for METs estimation.
The results were compared between devices. Accelerometer data from SGS were highly correlated with the accelerometer data from GT3X+ for all 3 axes, with a correlation ≥.89 for both the shaker test and treadmill test and ≥.70 for all daily activities, except for computer work. Our results for the classification of activity intensity levels, locomotion, sedentary, major body movement location, and individual activity recognition showed overall accuracies of 0.87, 1.00, 0.98, 0.85, and 0.64, respectively. The results were not significantly different between the SGS and GT3X+. Random forest model was the best model for METs estimation (root mean squared error of .71 and r-squared value of .50).
Our results suggest that a commercial brand smartwatch can be used in lieu of validated research grade activity monitors for individual activity recognition, major body movement location detection, activity intensity detection, and locomotion detection tasks.