Conference Paper

Comparison of Inertial Sensor Data from the Wrist and Mid-Lower Back During a 2-Minute Walk Test

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

Triaxial inertial sensor data was measured at the wrist and mid-lower back location during a 2-minute walk test. A Microsoft Band 2 was used to acquire gait data at the wrist and a Samsung Galaxy S5 smartphone was placed at the posterior pelvis. Accelerometer and gyroscope signals did not correlate well between the two locations. Important gait parameters at the wrist may lead to a better understanding of human locomotion profile due to periodic trends observed in wrist data. Turn detection accuracy was 86.0±23.1% and 78.3±20.2% for the MSB2 and smartphone, respectively. Turn detection showed that the smartphone y-axis gyroscope had more distinct peaks. Turn peak analysis showed that normalized peak heights were in agreement and MSB2 had a longer turn time.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... Emerging gait analysis techniques use embedded sensors in wearables and offer practical modalities for gait monitoring, human activity recognition (HAR), and prosthesis and orthosis control. Electronic sensors such as strain gauges [21], pressure sensors [23], electromyography (EMG) [17], force sensitive resistors [24][25][26], goniometers [27,28], and IMU [10,12,[23][24][25][26][29][30][31][32][33][34] can be combined to give highly accurate and real-time gait monitoring. ...
Article
Full-text available
Background Functionality and versatility of microprocessor-controlled stance-control knee-ankle-foot orthoses (M-SCKAFO) are dictated by their embedded control systems. Proper gait phase recognition (GPR) is required to enable these devices to provide sufficient knee-control at the appropriate time, thereby reducing the incidence of knee-collapse and fall events. Ideally, the M-SCKAFO sensor system would be local to the thigh and knee, to facilitate innovative orthosis designs that allow more flexibility for ankle joint selection and other orthosis components. We hypothesized that machine learning with local sensor signals from the thigh and knee could effectively distinguish gait phases across different walking conditions (i.e., surface levels, walking speeds) and that performance would improve with gait phase transition criteria (i.e., current states depend on previous states). Methods A logistic model decision tree (LMT) classifier was trained and tested (five-fold cross-validation) on gait data that included knee flexion angle, thigh-segment angular velocity, and thigh-segment acceleration. Twenty features were extracted from 0.1 s sliding windows for 30 able-bodied participants that walked on different surfaces (level, down-slope, up-slope, right cross-slope, left cross-slope) at a various walking speeds (self-paced (1.33 m/s, SD = 0.04 m/s), 0.8, 0.6, 0.4 m/s). The LMT-based GPR model was also tested with another validation set containing similar features and surfaces from 12 able-bodied volunteers at self-paced walking speeds (1.41 m/s, SD = 0.34 m/s). A “Transition Sequence Verification and Correction” (TSVC) algorithm was applied to correct for continuous class prediction and to improve GPR performance. Results The LMT had a tree size of 1643 with 822 leaf nodes, with a logistic regression model at each leaf node. The local sensor LMT-based GPR model identified loading response, push-off, swing, and terminal swing phases with overall classification accuracy of 98.38 for the initial training set (five-fold cross-validation) and 90.60% for the validation set. Applying TSVC increased classification accuracy to 98.72% for the initial training set and 98.61% for the validation set. Sensitivity, specificity, precision, F-score, and Matthew’s correlation coefficient results suggest strong evidence for the feasibility of an LMT-based GPR system for real-time orthosis control. Conclusions The novel machine learning GPR model that used sensor features local to the thigh and knee was viable for dynamic knee-ankle-foot orthosis-control. This highly accurate GPR model was generalizable when combined with TSVC. Our approach could reduce sensor system complexity as compared with other M-SCKAFO approaches, thereby enabling customizable advantages for end-users through modular unit orthosis designs.
Article
Full-text available
The 6-minute walk test (6MWT: the maximum distance walked in 6 minutes) is used by rehabilitation professionals as a measure of exercise capacity. Today's smartphones contain hardware that can be used for wearable sensor applications and mobile data analysis. A smartphone application can run the 6MWT and provide typically unavailable biomechanical information about how the person moves during the test. A new algorithm for a calibration-free 6MWT smartphone application was developed that uses the test's inherent conditions and smartphone accelerometer-gyroscope data to report the total distance walked, step timing, gait symmetry, and walking changes over time. This information is not available with a standard 6MWT and could help with clinical decision-making. The 6MWT application was evaluated with 15 able-bodied participants. A BlackBerry Z10 smartphone was worn on a belt at the mid lower back. Audio from the phone instructed the person to start and stop walking. Digital video was independently recorded during the trial as a gold-standard comparator. The average difference between smartphone and gold standard foot strike timing was 0.014 ± 0.015 s. The total distance calculated by the application was within 1 m of the measured distance for all but one participant, which was more accurate than other smartphone-based studies. These results demonstrated that clinically relevant 6MWT results can be achieved with typical smartphone hardware and a novel algorithm.
Article
Full-text available
In this paper a cumulant-based method for identification of gait using accelerometer data is presented. Acceleration data of three different walking speeds (slow, normal and fast) for each subject was acquired by the accelerometer embedded in cell phone which was attached to the person's hip. Data analysis was based on gait cycles that were detected first. Cumulants of order from 1 to 4 with different number of lags were calculated. Feature vectors for classification were built using dimension reduction on calculated cumulants by principal component analysis (PCA). The classification was accomplished by support vector machines (SVM) with radial basis kernel. According to portion of variance covered in the calculated principal components, different lengths of feature vectors were tested. Six healthy young subjects participated in the experiment. The average person recognition rate based on gait classification was 90.3±3.2%. A similarity measure for discerning different walking types of the same subject was also introduced using dimension reduction on accelerometer data by PCA.
Article
Full-text available
Gait analysis using wearable sensors is an inexpensive, convenient, and efficient manner of providing useful information for multiple health-related applications. As a clinical tool applied in the rehabilitation and diagnosis of medical conditions and sport activities, gait analysis using wearable sensors shows great prospects. The current paper reviews available wearable sensors and ambulatory gait analysis methods based on the various wearable sensors. After an introduction of the gait phases, the principles and features of wearable sensors used in gait analysis are provided. The gait analysis methods based on wearable sensors is divided into gait kinematics, gait kinetics, and electromyography. Studies on the current methods are reviewed, and applications in sports, rehabilitation, and clinical diagnosis are summarized separately. With the development of sensor technology and the analysis method, gait analysis using wearable sensors is expected to play an increasingly important role in clinical applications.
Article
Full-text available
As one of the most universal of all human activities, gait in the able-bodied has received considerable attention, but many aspects still need to be clarified. Symmetry or asymmetry in the actions of the lower extremities during walking and the possible effect of laterality on gait are two prevalent and controversial issues. The purpose of this study was to review the work done over the last few decades in demonstrating: (a) whether or not the lower limbs behave symmetrically during able-bodied gait; and (b) how limb dominance affects the symmetrical or asymmetrical behavior of the lower extremities. The literature reviewed shows that gait symmetry has often been assumed, to simplify data collection and analysis. In contrast, asymmetrical behavior of the lower limbs during able-bodied ambulation was addressed in numerous investigations and was found to reflect natural functional differences between the lower extremities. These functional differences were probably related to the contribution of each limb in carrying out the tasks of propulsion and control during able-bodied walking. In current debates on gait symmetry in able-bodied subjects, laterality has been cited as an explanation for the existence of functional differences between the lower extremities, although a number of studies do not support the hypothesis of a relationship between gait symmetry and laterality. Further investigation is needed to demonstrate functional gait asymmetry and its relationship to laterality, taking into consideration the biomechanical aspects of gait.
Article
Full-text available
In this work, 33 dimensional time-frequency domain features were developed and evaluated to detect five different human walking patterns from data acquired using a triaxial accelerometer attached at the waist above the iliac spine. 52 subjects were asked to walk on a flat surface along a corridor, walk up and down a flight of a stairway and walk up and down a constant gradient slope, in an unsupervised manner. Time-frequency domain features of acceleration data in anterior-posterior (AP), medio-lateral (ML) and vertical (VT) direction were developed. The acceleration signal in each direction was decomposed to six detailed signals at different wavelet scales by using the wavelet packet transform. The rms values and standard deviations of the decomposed signals at scales 5 to 2 corresponding to the 0.78-18.75 Hz frequency band were calculated. The energies in the 0.39-18.75 Hz frequency band of acceleration signal in AP, ML and VT directions were also computed. The back-end of the system was a multi-layer perceptron (MLP) Neural Networks (NNs) classifier. Overall classification accuracies of 88.54% and 92.05% were achieved by using a round robin (RR) and random frame selecting (RFS) train-test method respectively for the five walking patterns.
Conference Paper
Sensor-enabled smartphone's have become a mainstream platform for researchers due to their ability to collect and process large quantities of data, hence creating new opportunities for innovative applications. Yet, the limits in employing sensors to opportunistically detect human behaviors are not clear and deserve investigation. To this purpose, in this article, we discuss movement pattern recognition in day-by-day urban street behavior. As a case study, we restrict at recognizing situations when a pedestrian stops, crosses a street ruled by a traffic light; to do so we only use data coming from the accelerometer of the pedestrian's smartphone.
Symmetry and limb dominance in able-bodied gait
  • H Sadeghi
  • Allard
  • H Prince
  • Labelle