Visualization-Driven Time-Series Extraction from Wearable Systems Can Facilitate Differentiation of Passive ADL Characteristics among Stroke and Healthy Older Adults

Wearable technologies allow the measurement of unhindered activities of daily living (ADL) among patients who had a stroke in their natural settings. However, methods to extract meaningful information from large multi-day datasets are limited. This study investigated new visualization-driven time-series extraction methods for distinguishing activities from stroke and healthy adults. Fourteen stroke and fourteen healthy adults wore a wearable sensor at the L5/S1 position for three consecutive days and collected accelerometer data passively in the participant’s naturalistic environment. Data from visualization facilitated selecting information-rich time series, which resulted in classification accuracy of 97.3% using recurrent neural networks (RNNs). Individuals with stroke showed a negative correlation between their body mass index (BMI) and higher-acceleration fraction produced during ADL. We also found individuals with stroke made lower activity amplitudes than healthy counterparts in all three activity bands (low, medium, and high). Our findings show that visualization-driven time series can accurately classify movements among stroke and healthy groups using a deep recurrent neural network. This novel visualization-based time-series extraction from naturalistic data provides a physical basis for analyzing passive ADL monitoring data from real-world environments. This time-series extraction method using unit sphere projections of acceleration can be used by a slew of analysis algorithms to remotely track progress among stroke survivors in their rehabilitation program and their ADL abilities.

Correlations in sleeping patterns and circadian preference between spouses

Spouses may affect each other's sleeping behaviour. In 47,420 spouse-pairs from the UK Biobank, we found a weak positive phenotypic correlation between spouses for self-reported sleep duration (r=0.11; 95% CI=0.10, 0.12) and a weak inverse correlation for chronotype (diurnal preference) (r=-0.11; -0.12, -0.10), which replicated in up to 127,035 23andMe spouse-pairs. Using accelerometer data on 3,454 UK Biobank spouse-pairs, the correlation for derived sleep duration was similar to self-report (r=0.12; 0.09, 0.15). Timing of diurnal activity was positively correlated (r=0.24; 0.21, 0.27) in contrast to the inverse correlation for chronotype. In Mendelian randomization analysis, positive effects of sleep duration (mean difference=0.13; 0.04, 0.23 SD per SD) and diurnal activity (0.49; 0.03, 0.94) were observed, as were inverse effects of chronotype (-0.15; -0.26, -0.04) and snoring (-0.15; -0.27, -0.04). Findings support the notion that an individual's sleep may impact that of their partner, with implications for sleep health.

The Relation of Accelerometer-Measured Physical Activity and Serum Uric Acid Using the National Health and Nutrition Survey (NHANES) 2003–2004

Objective: Gout is a crystal-induced inflammatory arthritis caused by elevated uric acid. Physical activity has the potential to reduce serum uric acid (SUA), thus improving the disease burden of gout. In this study, we examined the association of objectively-measured physical activity and SUA.Methods: A cross-sectional study was conducted using survey, laboratory, and accelerometer data from the 2003–2004 National Health and Nutrition Examination Survey (NHANES). SUA concentrations (mg/dL) were obtained during an initial exam, and then physical activity (kCal/day) was measured with 7 days of ActiGraph accelerometry in participants (n = 3,475) representative of the ambulatory, non-institutionalized US civilian population. Regression, including restricted cubic splines, was used to assess the relation of physical activity and SUA in bivariate and adjusted models. Covariates included age, gender, race/ethnicity, alcohol use, body mass index, renal function, and urate-lowering therapy.Results: In the bivariate model, physical activity was correlated with SUA concentrations and included a non-linear component (p < 0.01). In the adjusted model, linear splines were employed with a node at the SUA nadir of 5.37mg/dL; this occurred at 703 kCal/day of physical activity. The association of physical activity and SUA was negative from 0 to 703 kCal/day (p = 0.07) and positive >703 kCal/day (p < 0.01 for the change in slope).Conclusion: Physical activity and SUA are associated in a non-linear fashion, with a minimum estimated SUA at 703 kCal/day of objectively-measured physical activity. These findings raise intriguing questions about the use of physical activity as a potential adjunctive therapy in patients with gout, and further interventional studies are needed to elucidate the effects of moderate intensity exercise on SUA concentrations.

Predicting continuous ground reaction forces from accelerometers during uphill and downhill running: a recurrent neural network solution

Background Ground reaction forces (GRFs) are important for understanding human movement, but their measurement is generally limited to a laboratory environment. Previous studies have used neural networks to predict GRF waveforms during running from wearable device data, but these predictions are limited to the stance phase of level-ground running. A method of predicting the normal (perpendicular to running surface) GRF waveform using wearable devices across a range of running speeds and slopes could allow researchers and clinicians to predict kinetic and kinematic variables outside the laboratory environment. Purpose We sought to develop a recurrent neural network capable of predicting continuous normal (perpendicular to surface) GRFs across a range of running speeds and slopes from accelerometer data. Methods Nineteen subjects ran on a force-measuring treadmill at five slopes (0°, ±5°, ±10°) and three speeds (2.5, 3.33, 4.17 m/s) per slope with sacral- and shoe-mounted accelerometers. We then trained a recurrent neural network to predict normal GRF waveforms frame-by-frame. The predicted versus measured GRF waveforms had an average ± SD RMSE of 0.16 ± 0.04 BW and relative RMSE of 6.4 ± 1.5% across all conditions and subjects. Results The recurrent neural network predicted continuous normal GRF waveforms across a range of running speeds and slopes with greater accuracy than neural networks implemented in previous studies. This approach may facilitate predictions of biomechanical variables outside the laboratory in near real-time and improves the accuracy of quantifying and monitoring external forces experienced by the body when running.

Daily Activity Recognition and Tremor Quantification from Accelerometer Data for Patients with Essential Tremor Using Stacked Denoising Autoencoders

Human activity recognition (HAR) has received more and more attention, which is able to play an important role in many fields, such as healthcare and intelligent home. Thus, we have discussed an application of activity recognition in the healthcare field in this paper. Essential tremor (ET) is a common neurological disorder that can make people with this disease rise involuntary tremor. Nowadays, the disease is easy to be misdiagnosed as other diseases. We have combined the essential tremor and activity recognition to recognize ET patients’ activities and evaluate the degree of ET for providing an auxiliary analysis toward disease diagnosis by utilizing stacked denoising autoencoder (SDAE) model. Meanwhile, it is difficult for model to learn enough useful features due to the small behavior dataset from ET patients. Thus, resampling techniques are proposed to alleviate small sample size and imbalanced samples problems. In our experiment, 20 patients with ET and 5 healthy people have been chosen to collect their acceleration data for activity recognition. The experimental results show the significant result on ET patients activity recognition and the SDAE model has achieved an overall accuracy of 93.33%. What’s more, this model is also used to evaluate the degree of ET and has achieved the accuracy of 95.74%. According to a set of experiments, the model we used is able to acquire significant performance on ET patients activity recognition and degree of tremor assessment.

Robust Human Activity Recognition by Integrating Image and Accelerometer Sensor Data Using Deep Fusion Network

Studies on deep-learning-based behavioral pattern recognition have recently received considerable attention. However, if there are insufficient data and the activity to be identified is changed, a robust deep learning model cannot be created. This work contributes a generalized deep learning model that is robust to noise not dependent on input signals by extracting features through a deep learning model for each heterogeneous input signal that can maintain performance while minimizing preprocessing of the input signal. We propose a hybrid deep learning model that takes heterogeneous sensor data, an acceleration sensor, and an image as inputs. For accelerometer data, we use a convolutional neural network (CNN) and convolutional block attention module models (CBAM), and apply bidirectional long short-term memory and a residual neural network. The overall accuracy was 94.8% with a skeleton image and accelerometer data, and 93.1% with a skeleton image, coordinates, and accelerometer data after evaluating nine behaviors using the Berkeley Multimodal Human Action Database (MHAD). Furthermore, the accuracy of the investigation was revealed to be 93.4% with inverted images and 93.2% with white noise added to the accelerometer data. Testing with data that included inversion and noise data indicated that the suggested model was robust, with a performance deterioration of approximately 1%.

The Dilemma of Analyzing Physical Activity and Sedentary Behavior with Wrist Accelerometer Data: Challenges and Opportunities

Physical behaviors (e.g., physical activity and sedentary behavior) have been the focus among many researchers in the biomedical and behavioral science fields. The recent shift from hip- to wrist-worn accelerometers in these fields has signaled the need to develop novel approaches to process raw acceleration data of physical activity and sedentary behavior. However, there is currently no consensus regarding the best practices for analyzing wrist-worn accelerometer data to accurately predict individuals’ energy expenditure and the times spent in different intensities of free-living physical activity and sedentary behavior. To this end, accurately analyzing and interpreting wrist-worn accelerometer data has become a major challenge facing many clinicians and researchers. In response, this paper attempts to review different methodologies for analyzing wrist-worn accelerometer data and offer cutting edge, yet appropriate analysis plans for wrist-worn accelerometer data in the assessment of physical behavior. In this paper, we first discuss the fundamentals of wrist-worn accelerometer data, followed by various methods of processing these data (e.g., cut points, steps per minute, machine learning), and then we discuss the opportunities, challenges, and directions for future studies in this area of inquiry. This is the most comprehensive review paper to date regarding the analysis and interpretation of free-living physical activity data derived from wrist-worn accelerometers, aiming to help establish a blueprint for processing wrist-derived accelerometer data.