scholarly journals Random forest algorithms for recognizing daily life activities using plantar pressure information: a smart-shoe study

PeerJ ◽  
2020 ◽  
Vol 8 ◽  
pp. e10170
Author(s):  
Dian Ren ◽  
Nathanael Aubert-Kato ◽  
Emi Anzai ◽  
Yuji Ohta ◽  
Julien Tripette
Keyword(s):  
Random Forest ◽  
Activity Recognition ◽  
Plantar Pressure ◽  
Flat Surface ◽  
Daily Life ◽  
Smart Devices ◽  
Pressure Data ◽  
Process Data ◽  
Daily Life Activities ◽  
Daily Physical Activities

Background 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. Method 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. Results 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. Conclusion 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.

Daily-life activities and in-shoe forefoot plantar pressure in patients with diabetes

2007 ◽  
Vol 77 (2) ◽  
pp. 203-209 ◽  
Author(s):  
Nick A. Guldemond ◽  
Pieter Leffers ◽  
Antal P. Sanders ◽  
Nicolaas C. Schaper ◽  
Fred Nieman ◽  
...  
Keyword(s):  
Plantar Pressure ◽  
Daily Life ◽  
Daily Life Activities ◽  
Patients With Diabetes

scholarly journals Analyzing the Effectiveness and Contribution of Each Axis of Tri-Axial Accelerometer Sensor for Accurate Activity Recognition

Sensors ◽  
2020 ◽  
Vol 20 (8) ◽  
pp. 2216 ◽  
Author(s):  
Abdul Rehman Javed ◽  
Muhammad Usman Sarwar ◽  
Suleman Khan ◽  
Celestine Iwendi ◽  
Mohit Mittal ◽  
...  
Keyword(s):  
Physical Activity ◽  
Activity Recognition ◽  
Physical Activities ◽  
Smart Devices ◽  
Accelerometer Sensor ◽  
Daily Life Activities ◽  
Adaptive Services ◽  
Physical Activity Recognition ◽  
Active Research ◽  
Mlp Classifier

Recognizing human physical activities from streaming smartphone sensor readings is essential for the successful realization of a smart environment. Physical activity recognition is one of the active research topics to provide users the adaptive services using smart devices. Existing physical activity recognition methods lack in providing fast and accurate recognition of activities. This paper proposes an approach to recognize physical activities using only2-axes of the smartphone accelerometer sensor. It also investigates the effectiveness and contribution of each axis of the accelerometer in the recognition of physical activities. To implement our approach, data of daily life activities are collected labeled using the accelerometer from 12 participants. Furthermore, three machine learning classifiers are implemented to train the model on the collected dataset and in predicting the activities. Our proposed approach provides more promising results compared to the existing techniques and presents a strong rationale behind the effectiveness and contribution of each axis of an accelerometer for activity recognition. To ensure the reliability of the model, we evaluate the proposed approach and observations on standard publicly available dataset WISDM also and provide a comparative analysis with state-of-the-art studies. The proposed approach achieved 93% weighted accuracy with Multilayer Perceptron (MLP) classifier, which is almost 13% higher than the existing methods.

scholarly journals Tracking and Monitoring Fitness of Athletes Using IoT Enabled Wearables for Activity Recognition and Random Forest Algorithm for Performance Prediction

2021 ◽  
pp. 72-86
Author(s):  
Krishna Prasad K. ◽  
P. S. Aithal ◽  
Geetha Poornima K. ◽  
Vinayachandra
Keyword(s):  
Random Forest ◽  
Conceptual Model ◽  
Activity Recognition ◽  
Performance Prediction ◽  
Emotional Health ◽  
Wearable Devices ◽  
Physical Activities ◽  
Smart Devices ◽  
Related Services ◽  
And Performance

Purpose: The progression in technology is made the best use of in every field. Sports analytics is an essential sector that has gained importance in this technology-driven era. It is used to determine the hidden relationships among different quantitative parameters that affect the performance of athletes. This type of analysis requires a large amount of data to be stored periodically. Cloud acts as a scalable centralized repository that can store the massive data essential for analysis purpose. From the technological perspective there are numerous wearable activity tracking devices, which will be able to provide feedback of physical activities. With the help of random forest (RF) algorithm it is possible to classify huge datasets to perform predictions. In this paper, different smart devices that can be used to measure physical activity, use of RF algorithm for converting data obtained from smart devices into knowledge are explored. A conceptual model that uses wearable devices for tracking and monitoring and RF algorithm to predict the performance is suggested. Methodology: The study was conducted by referring to scholarly documents available online and by referring to websites of companies offering healthcare and sports related services. A conceptual model is developed based on the theoretical perception that incorporates the components needed for measuring the physical activities to predict the performance of athletes. Findings/Result: In this paper the proposed system contains four major activities as Capture, Store, Analyze, and Predict. The model considers use of IoT-enabled wearable devices to measure the physical activities of athletes and the information collected will in turn be used to analyze predict their performance and suggest them how to increase the chances of winning. However, the outcome of a game does not only depend upon the PA of athletes. It depends also upon the physical, mental, emotional health, nutrition and many other factors. Originality: In this paper, a theoretical model is deduced to integrate IoT and RF Algorithm to track and monitor fitness of athletes using wearables for activity recognition and performance prediction. Paper Type: Conceptual Paper

Mobile phone apps’ support for people with mild dementia: A systematic review (Preprint)

2019 ◽  
Author(s):  
Leona Cilar ◽  
Lucija Gosak ◽  
Amanda Briggs ◽  
Klavdija Čuček Trifkovič ◽  
Tracy McClelland ◽  
...  
Keyword(s):  
Mobile Phone ◽  
Thematic Analysis ◽  
Daily Life ◽  
General Term ◽  
Cognitive Domain ◽  
Mild Dementia ◽  
Daily Life Activities ◽  
Human Needs ◽  
People With Dementia ◽  
System Usability Scale

BACKGROUND Dementia is a general term for various disorders characterized by memory impairment and loss of at least one cognitive domain. People with dementia are faced with different difficulties in their daily life activities (DLA). With the use of modern technologies, such as mobile phone apps – often called health apps, their difficulties can be alleviated. OBJECTIVE The aim of this paper was to systematically search, analyze and synthetize mobile phone apps designed to support people with mild dementia in daily life activities in two apps bases: Apple App Store and Google Play Store. METHODS A search was conducted in May 2019 following PRISMA recommendations. Results were analyzed and displayed as tables and graphs. Results were synthetized using thematic analysis which was conducted from 14 components, based on human needs for categorized nursing activities. Mobile phone apps were assessed for quality using the System Usability Scale. RESULTS A total of 15 mobile phone apps were identified applying inclusion and exclusion criteria. Five major themes were identified with thematic analysis: multi-component DLA, communication and feelings, recreation, eating and drinking, and movement. Most of the apps (73%) of the apps were not mentioned in scientific literature. CONCLUSIONS There are many mobile phone apps available in mobile phone markets for the support for people with mild dementia; yet only a few of them are focused on challenges in daily life activities. Most of the available apps were not evaluated nor assessed for quality.

scholarly journals Plantar Pressure Differences Between Nordic Walking Techniques

2017 ◽  
Vol 57 (1) ◽  
pp. 221-231 ◽  
Author(s):  
Alberto Encarnación-Martínez ◽  
Ángel Gabriel Lucas-Cuevas ◽  
Pedro Pérez-Soriano ◽  
Ruperto Menayo ◽  
Gemma María Gea-García
Keyword(s):  
Plantar Pressure ◽  
Repeated Measures ◽  
Walking Speed ◽  
Lower Limbs ◽  
Pressure Data ◽  
Nordic Walking ◽  
Risk Of Injury ◽  
Increased Risk ◽  
Practical Conclusion ◽  
Physiological Benefits

AbstractHigh plantar pressure has been associated with increased risk of injury. The characteristics of each physical activity determine the load on the lower limbs. The influence of Nordic Walking (NW) technique on plantar pressure is still unknown. The aim of this study was to analyze the differences between plantar pressure during NW with the Diagonal technique (DT) versus Alpha technique (AT) and compare them with the pressure obtained during normal walking (W). The normality and sphericity of the plantar pressure data were checked before performing a two-way repeated measures ANOVA in order to find differences between speeds (preferred, fast) and the gait (NW, W) as within-subject factors. Then, a t-test for independent measures was used to identify the specific differences between NW techniques. The strength of the differences was calculated by means of the effect size (ES). The results demonstrated that during NW with AT at preferred speed the pressure was lower under the Calcaneus, Lateral Metatarsal and Toes compared to the DT group (p = 0.046, ES = 1.49; p = 0.015, ES = 1.44; p = 0.040, ES = 1.20, respectively). No differences were found at the fast speed (p > 0.05). Besides the increase in walking speed during NW (p < 0.01), both technique groups showed lower pressure during NW compared to W under the Hallux and Central Metatarsal heads (F = 58.321, p = 0.000, ES = 2.449; F = 41.917, p = 0.012, ES = 1.365, respectively). As a practical conclusion, the AT technique may be the most effective of the NW techniques at reducing plantar pressure while allowing NW practitioners to achieve the physiological benefits of NW.

Sign in / Sign up

Export Citation Format

Share Document