Dimensional reduction of balance parameters in risk of falling evaluation using a minimal number of force-sensitive resistors.

Purpose. As the instrumented insole is available for a wide commercial range in the retail trade, this study aims to reduce its overall cost using fewer sensors by carrying out an effective risk of falling evaluation. Methods. We compared the effect of reducing balance parameters using four and three force-sensing resistors (FSRs) of an instrumented insole. Data were previously collected among elderly participants during a Timed Up and Go (TUG) test. Results. While reducing the number of balance parameters, during sit-to-stand and stand-to-sit activities, the risk scores using four FSRs were not significantly different compared with three FSRs. Parameter reduction did not show any significant loss of information among the study population using four FSRs. For certain configurations of three FSRs, a significant effect of information loss was found in the study participants, revealing the importance of investigating the sensor locations in the process. Conclusions. We conclude that it is feasible to estimate a risk index during a TUG test not only after reducing the number of needed sensing units from four to three FSRs but also after reducing the number of balance parameters. The three FSRs should be located at strategic positions to avoid a significant loss of information.

Image Generation for 2D-CNN Using Time-Series Signal Features from Foot Gesture Applied to Select Cobot Operating Mode.

Advances in robotics are part of reducing the burden associated with manufacturing tasks in workers. For example, the cobot could be used as a "third-arm" during the assembling task. Thus, the necessity of designing new intuitive control modalities arises. This paper presents a foot gesture approach centered on robot control constraints to switch between four operating modalities. This control scheme is based on raw data acquired by an instrumented insole located at a human's foot. It is composed of an inertial measurement unit (IMU) and four force sensors. Firstly, a gesture dictionary was proposed and, from data acquired, a set of 78 features was computed with a statistical approach, and later reduced to 3 via variance analysis ANOVA. Then, the time series collected data were converted into a 2D image and provided as an input for a 2D convolutional neural network (CNN) for the recognition of foot gestures. Every gesture was assimilated to a predefined cobot operating mode. The offline recognition rate appears to be highly dependent on the features to be considered and their spatial representation in 2D image. We achieve a higher recognition rate for a specific representation of features by sets of triangular and rectangular forms. These results were encouraging in the use of CNN to recognize foot gestures, which then will be associated with a command to control an industrial robot.

Risk of Falling in a Timed Up and Go Test Using an UWB Radar and an Instrumented Insole.

Previously, studies reported that falls analysis is possible in the elderly, when using wearable sensors. However, these devices cannot be worn daily, as they need to be removed and recharged from time-to-time due to their energy consumption, data transfer, attachment to the body, etc. This study proposes to introduce a radar sensor, an unobtrusive technology, for risk of falling analysis and combine its performance with an instrumented insole. We evaluated our methods on datasets acquired during a Timed Up and Go (TUG) test where a stride length (SL) was computed by the insole using three approaches. Only the SL from the third approach was not statistically significant (p = 0.2083 > 0.05) compared to the one provided by the radar, revealing the importance of a sensor location on human body. While reducing the number of force sensors (FSR), the risk scores using an insole containing three FSRs and y-axis of acceleration were not significantly different (p > 0.05) compared to the combination of a single radar and two FSRs. We concluded that contactless TUG testing is feasible, and by supplementing the instrumented insole to the radar, more precise information could be available for the professionals to make accurate decision.

Low-Rank and Sparse Recovery of Human Gait Data.

Due to occlusion or detached markers, information can often be lost while capturing human motion with optical tracking systems. Based on three natural properties of human gait movement, this study presents two different approaches to recover corrupted motion data. These properties are used to define a reconstruction model combining low-rank matrix completion of the measured data with a group-sparsity prior on the marker trajectories mapped in the frequency domain. Unlike most existing approaches, the proposed methodology is fully unsupervised and does not need training data or kinematic information of the user. We evaluated our methods on four different gait datasets with various gap lengths and compared their performance with a state-of-the-art approach using principal component analysis (PCA). Our results showed recovering missing data more precisely, with a reduction of at least 2 mm in mean reconstruction error compared to the literature method. When a small number of marker trajectories is available, our findings showed a reduction of more than 14 mm for the mean reconstruction error compared to the literature approach.

Wearable Devices for Classification of Inadequate Posture at Work Using Neural Networks.

Inadequate postures adopted by an operator at work are among the most important risk factors in Work-related Musculoskeletal Disorders (WMSDs). Although several studies have focused on inadequate posture, there is limited information on its identification in a work context. The aim of this study is to automatically differentiate between adequate and inadequate postures using two wearable devices (helmet and instrumented insole) with an inertial measurement unit (IMU) and force sensors. From the force sensors located inside the insole, the center of pressure (COP) is computed since it is considered an important parameter in the analysis of posture. In a first step, a set of 60 features is computed with a direct approach, and later reduced to eight via a hybrid feature selection. A neural network is then employed to classify the current posture of a worker, yielding a recognition rate of 90%. In a second step, an innovative graphic approach is proposed to extract three additional features for the classification. This approach represents the main contribution of this study. Combining both approaches improves the recognition rate to 95%. Our results suggest that neural network could be applied successfully for the classification of adequate and inadequate posture.

Comparing auditory, visual and vibrotactile cues in individuals with Parkinson's disease for reducing risk of falling over different types of soil.

INTRODUCTION: Several researchers have demonstrated the positive benefits of auditory and visual cueing in the gait improvements among individuals with Parkinson's disease (PD). However, few studies have evaluated the role of vibrotactile cueing when compared to auditory and visual cueing. This paper compares how these stimuli affect the risk of falling while walking on six types of soil (concrete, sand, parquet, broken stone, two types of carpet). METHODS: An instrumented Timed Up and Go (iTUG) test served to evaluate how audio, visual and vibrotactile cueing can affect the risk of falling of elderly. This pilot study proposes 12 participants with PD (67.7 ± 10.07 years) and nine age-matched controls (66.8 ± 8.0 years). Both groups performed the iTUG test with and without cueing. The cueing frequency was set at 10% above the cadence computed at the lower risk level of falling (walking over the concrete). A computed risk of falling (ROFA) index has been compared to the TUG time (total TUG duration). RESULTS: The index for evaluating the risk of falling appears to have a good reliability (ICC > 0.88) in this pilot study. In addition, the minimal detectable change (MDC) suggests that the proposed index could be more sensitive to the risk of falling variation compared to the TUG time. Moreover, while using the cueing, observed results suggest a significant decrease in the computed risk of falling compared to 'without cueing' for most of types of soil, especially for deformable soils, which can lead to falls. CONCLUSION: When compared to other cueing, it seems that audio could be a better neurofeedback for reducing the risk of falling over different walking surfaces, which represent important risk factors for persons with gait disorder or lost functional autonomy.

Use of an Enactive Insole for Reducing the Risk of Falling on Different Types of Soil Using Vibrotactile Cueing for the Elderly.

BACKGROUND: Our daily activities imply displacements on various types of soil. For persons with gait disorder or losing functional autonomy, walking on some types of soil could be challenging because of the risk of falling it represents. METHODS: In this paper, we present, in a first part, the use of an enactive shoe for an automatic differentiation of several types of soil. In a second part, using a second improved prototype (an enactive insole), twelve participants with Parkinson's disease (PD) and nine age-matched controls have performed the Timed Up and Go (TUG) test on six types of soil with and without cueing. The frequency of the cueing was set at 10% above the cadence computed at the lower risk of falling (walking over the concrete). Depending on the cadence computed at the lower risk, the enactive insole activates a vibrotactile cueing aiming to improve gait and balance control. Finally, a risk index is computed using gait parameters in relation to given type of soil. RESULTS: The frequency analysis of the heel strike vibration allows the differentiation of various types of soil. The risk computed is associated to an appropriate rhythmic cueing in order to improve balance and gait impairment. The results show that a vibrotactile cueing could help to reduce the risk of falling. CONCLUSIONS: Firstly, this paper demonstrates the feasibility of reducing the risk of falling while walking on different types of soil using vibrotactile cueing. We found a significant difference and a significant decrease in the computed risks of falling for most of types of soil especially for deformable soils which can lead to fall. Secondly, heel strike provides an approximation of the impulse response of the soil that can be analyzed with time and frequency-domain modeling. From these analyses, an index is computed enabling differentiation the types of soil.

Home-Based Risk of Falling Assessment Test Using a Closed-Loop Balance Model.

The aim of this study is to improve and facilitate the methods used to assess risk of falling at home among older people through the computation of a risk of falling in real time in daily activities. In order to increase a real time computation of the risk of falling, a closed-loop balance model is proposed and compared with One-Leg Standing Test (OLST). This balance model allows studying the postural response of a person having an unpredictable perturbation. Twenty-nine volunteers participated in this study for evaluating the effectiveness of the proposed system which includes seventeen elder participants: ten healthy elderly ( 68.4 ±5.5 years), seven Parkinson's disease (PD) subjects ( 66.28 ±8.9 years), and twelve healthy young adults ( 28.27 ±3.74 years). Our work suggests that there is a relationship between OLST score and the risk of falling based on center of pressure measurement with four low cost force sensors located inside an instrumented insole, which could be predicted using our suggested closed-loop balance model. For long term monitoring at home, this system could be included in a medical electronic record and could be useful as a diagnostic aid tool.