Search

Consumer-grade wearable activity devices such as Fitbits are increasingly being used in research settings to promote physical activity (PA) due to their low-cost and widespread popularity. However, Fitbit-derived measures of activity intensity are consistently reported to be less accurate than intensity estimates obtained from research-grade accelerometers (i.e., ActiGraph). As such, the potential for using a Fitbit to measure PA intensity within research contexts remains limited. This study aims to model ActiGraph-based intensity estimates from the validated Freedson vector magnitude (VM3) algorithm using measures of steps, metabolic equivalents, and intensity levels obtained from Fitbit. Minute-level data collected from 19 subjects, who concurrently wore the ActiGraph GT3X and Fitbit Flex devices for an average of 1.8 weeks, were used to generate the model. After testing several modeling methods, a naïve Bayes classifier was chosen based on the lowest achieved error rate. Overall, the model reduced Fitbit to ActiGraph errors from 19.97% to 16.32%. Moreover, the model reduced misclassification of Fitbit-based estimates of moderate-to-vigorous physical activity (MVPA) by 40%, eliminating a statistically significant difference between MVPA estimates derived from ActiGraph and Fitbit. Study findings support the general utility of the model for measuring MVPA with the Fitbit Flex in place of the more costly ActiGraph GT3X accelerometer for young healthy adults.

In this study we examined changes to the human gut microbiome resulting from an eight-week intervention of either cardiorespiratory exercise (CRE) or resistance training exercise (RTE). Twenty-eight subjects (21 F, aged 18&ndash, 26) were recruited for our CRE study and 28 subjects (17 F, 33) were recruited for our RTE study. Fecal samples for gut microbiome profiling were collected twice weekly during the pre-intervention phase (three weeks), intervention phase (eight weeks), and post-intervention phase (three weeks). Pre/post VO2max, three repetition maximum (3RM), and body composition measurements were conducted. Heart rate ranges for CRE were determined by subjects&rsquo, initial VO2max test. RTE weight ranges were established by subjects&rsquo, initial 3RM testing for squat, bench press, and bent-over row. Gut microbiota were profiled using 16S rRNA gene sequencing. Microbiome sequence data were analyzed with QIIME 2. CRE resulted in initial changes to the gut microbiome which were not sustained through or after the intervention period, while RTE resulted in no detectable changes to the gut microbiota. For both CRE and RTE, we observe some evidence that the baseline microbiome composition may be predictive of exercise gains. This work suggests that the human gut microbiome can change in response to a new exercise program, but the type of exercise likely impacts whether a change occurs. The changes observed in our CRE intervention resemble a disturbance to the microbiome, where an initial shift is observed followed by a return to the baseline state. More work is needed to understand how sustained changes to the microbiome occur, resulting in differences that have been reported in cross sectional studies of athletes and non-athletes.

The purpose of this paper is to present a method to detect bike trail conditions while displaying real-time data on a mobile application. Using a Global Positioning System (GPS) unit, microprocessor and multiple accelerometers built on the concept of Internet of Things (IoT), the entire device employs real-time data to collect, analyze, and identify bumps and cracks in public infrastructure. The main purpose of this research is aimed at helping the government evaluate the severity of road conditions to better assist in strategic decision making about maintaining or repairing damaged infrastructural areas as well as to notify pedestrians of potential bumps or cracks along their route. The authors propose an algorithm that sets a window range to vertically process the raw accelerometer data to accurately distinguish bumps. By integrating this algorithm onto a mobile application, the data collected by this program can be shared between users to help avoid potential risks or dangerous road conditions and to assist in the evaluation of infrastructural deterioration.

This paper presents a method of stride identification, extraction, and analysis of data sets of time-series contact force data for ambulating subjects both with and without Parkinson's disease (PD). This method has been made robust with the use of seeded K-Means clustering, fast Fourier transformation (FFT) spectral analysis, and minimum window size rejection. These methods combine to produce well selected strides of active walking data. We are able to calculate quality of walking measures of stride duration, stance duration (as percent of gait cycle - %GC), swing duration (%GC), time to maximum heel force (%GC), time to maximum toe force (%GC), time spent in heel contact (%GC), and time spent in toe contact (%GC).

Fitbit devices are one of the most popular wearable activity monitors in the consumer market. They are considerably cheaper than many of their clinical grade counterparts. However, they utilize proprietary algorithms for estimation of physical activity (PA). This study aims to model the measures of PA as reported by the ActiGraph GT3X using Fitbit measures of steps, METs, and intensity level. Such a model relating the Fitbit to what would have been reported by the ActiGraph could enable researchers to use the Fitbit instead of the ActiGraph in some applications, thus reducing cost or increasing the number of subjects involved in a study. This paper describes a study in which a model of the Freedson VM3 physical activity classification was constructed that uses measures from the Fitbit device instead of the typically provided ActiGraph vector magnitude. The data from 19 subjects, who concurrently wore both the ActiGraph and Fitbit devices for an average of 1.8 weeks, was used to generate the minute level based model. Several modeling methods were tested; a naive Bayes classifier was chosen based on the lowest achieved error rate. That model reduces overall Fitbit to Acti-Graph errors from 19.97% to 16.32%, a notable improvement. More importantly, it reduces the errors in moderate to vigorous physical activity levels by 40%, eliminating a statistically significant difference between MVPA estimates provided by the Freedson VM3 and Fitbit Intensity scores. This justifies use of the Fitbit device in place of an ActiGraph device in some large scale studies, especially those where MVPA estimates are of importance.

Human activity can be measured through identification of bouts of activity. The Freedson cut point method used by ActiGraph has become one very common and well accepted standard for estimating times of continuous moderate to vigorous physical activity (MVPA). However, such methods do not directly apply to other data sources such as the Fitbit Flex, a wrist worn wireless pedometer. In previous research by the authors, a model was presented to improve the estimates of physical activity (PA) level in the Fitbit devices. This paper considers the estimates of activity bouts, building on the modeled PA level from the Fitbit Flex as compared to the results from the ActiGraph GT3X. The purpose of this paper is to compare the “gold standard” ActiGraph to modeled Fitbit Freedson methods and to establish normative values of expected errors in bout detection between the two devices and methods, both of which are proxy methods aimed at measuring actual physical activity levels. Here we compare bout identification using three measures, the ActiGraph Freedson method, Fitbit Intensity Score, and the modeled Fitbit Freedson using three different outcomes. First, we compare a baseline of per subject per day number and duration of bouts from an ActiGraph GT3X to the results found from using the same methods on the Intensity Score reported by Fitbit and the modeled Fitbit Freedson method. Next, we compare the difference in duration of bouts identified in each data source matched according to similar start and end times. Finally, we compare the bouts found from the three methods to bouts identified in a self report diary.

Identifying noninvasive treatments to alleviate the symptoms of Parkinson's disease (PD) is important to improving the quality of life for those with PD. Several studies have explored the effects of visual, auditory, and vibratory cueing to improve gait in PD patients. Here, we present a wireless vibratory feedback system, called the PDShoe, and an associated intervention study with four subjects. The PDShoe was used on two control subjects, one subject with PD who experienced freezing of gait (FOG), and one subject with PD with an implanted deep brain stimulator (DBS). This short intervention study showed statistically significant improvements in peak heel pressure timing, peak toe pressure timing, time on the heel sensor, and stance to swing ratio after just one week of twice-daily therapy. Thus, step-synchronized vibration applied to the feet of patients with PD may be an effective way to improve gait in those subjects.

BackgroundWearable activity monitors such as Fitbit enable users to track various attributes of their physical activity (PA) over time and have the potential to be used in research to promote and measure PA behavior. However, the measurement accuracy of Fitbit in absolute free-living conditions is largely unknown. ObjectiveTo examine the measurement congruence between Fitbit Flex and ActiGraph GT3X for quantifying steps, metabolic equivalent tasks (METs), and proportion of time in sedentary activity and light-, moderate-, and vigorous-intensity PA in healthy adults in free-living conditions. MethodsA convenience sample of 19 participants (4 men and 15 women), aged 18-37 years, concurrently wore the Fitbit Flex (wrist) and ActiGraph GT3X (waist) for 1- or 2-week observation periods (n=3 and n=16, respectively) that included self-reported bouts of daily exercise. Data were examined for daily activity, averaged over 14 days and for minutes of reported exercise. Average day-level data included steps, METs, and proportion of time in different intensity levels. Minute-level data included steps, METs, and mean intensity score (0 = sedentary, 3 = vigorous) for overall reported exercise bouts (N=120) and by exercise type (walking, n=16; run or sports, n=44; cardio machine, n=20). ResultsMeasures of steps were similar between devices for average day- and minute-level observations (all P values > .05). Fitbit significantly overestimated METs for average daily activity, for overall minutes of reported exercise bouts, and for walking and run or sports exercises (mean difference 0.70, 1.80, 3.16, and 2.00 METs, respectively; all P values < .001). For average daily activity, Fitbit significantly underestimated the proportion of time in sedentary and light intensity by 20% and 34%, respectively, and overestimated time by 3% in both moderate and vigorous intensity (all P values < .001). Mean intensity scores were not different for overall minutes of exercise or for run or sports and cardio-machine exercises (all P values > .05). ConclusionsFitbit Flex provides accurate measures of steps for daily activity and minutes of reported exercise, regardless of exercise type. Although the proportion of time in different intensity levels varied between devices, examining the mean intensity score for minute-level bouts across different exercise types enabled interdevice comparisons that revealed similar measures of exercise intensity. Fitbit Flex is shown to have measurement limitations that may affect its potential utility and validity for measuring PA attributes in free-living conditions.

This paper discusses the design of a new, minimally constraining, passively supported gait training exoskeleton known as ALEX II. This device builds on the success and extends the features of the ALEX I device developed at the University of Delaware. Both ALEX (Active Leg EXoskeleton) devices have been designed to supply a controllable torque to a subject's hip and knee joint. The current control strategy makes use of an assist-as-needed algorithm. Following a brief review of previous work motivating this redesign, we discuss the key mechanical features of the new ALEX device. A short investigation was conducted to evaluate the effectiveness of the control strategy and impact of the exoskeleton on the gait of six healthy subjects. This paper concludes with a comparison between the subjects' gait both in and out of the exoskeleton.

This paper outlines how a control moment gyroscope can be used to generate haptic torque feedback while minimizing the effects of a constrained gimbal workspace. We present the design of the iTorqU 2.1 and discuss how it compares to previously developed systems. We then detail the control algorithms we have developed for both transparency and torque output modes. The prescribed transparency controller is typical in design, but the torque output algorithm is novel to this type of haptic device. It makes use of a series of position-p-at-time-t commands, which we call packets. Five packet designs were considered in this research, but we have included only the most important three in this paper. While this research deals with torque feedback, it ultimately presents a method for working with devices that are limited by the need for continuous reset to a home position before subsequent outputs.

This paper discusses the design and operation of the iTorqU 2.0, an ungrounded, handheld torque feedback device for haptic applications. Based upon the gyroscopic effect, the iTorqU 2.0 uses a metal flywheel inside of a two-axis actuated gimbal to create directional torques that are applied to the user's hand. The coupling of angular velocity and angular momentum creates a torque that is orthogonal to the two input angular velocities, giving the user the impression that their hand is being twisted in free air. Following a review of prior work in the field of ungrounded torque feedback devices, we first present our preliminary prototype, the iTorqU 1.0. Building on empirical observations and user feedback from a public demonstration, we revised and augmented this design to create the iTorqU 2.0. This paper covers the major mechanical, electrical, and controls design considerations that went into creating the iTorqU 2.0, along with an analysis of its torque output capabilities.

Subjects with Parkinson's disease (PD) often have trouble with ambulation. Some research has shown that auxiliary cueing in the form of vision, audio, or vibration can improve the gait of PD patients. We have developed a new vibratory feedback shoe, known as the PDShoe, which builds on existing research. This device can modulate both frequency and amplitude of feedback for the wearer. It is untethered, and thus can be worn during daily activities. Pressure and tactor status data are transmitted wirelessly over a personal area network to a notebook computer. This computer can also control the tactor actuation and stimulation frequency. This paper describes the details of design and construction of the PDShoe. A preliminary evaluation with four Parkinson's disease subjects and two healthy subjects is included to show the usability of the device.