scholarly journals Deep Learning for Classifying Physical Activities from Accelerometer Data

Sensors ◽  
2021 ◽  
Vol 21 (16) ◽  
pp. 5564
Author(s):  
Vimala Nunavath ◽  
Sahand Johansen ◽  
Tommy Sandtorv Johannessen ◽  
Lei Jiao ◽  
Bjørge Herman Hansen ◽  
...  
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Activity Patterns ◽  
Physical Activities ◽  
Accelerometer Data ◽  
Feed Forward Neural Network ◽  
Physical Movement ◽  
Colon Cancers ◽  
Deep Recurrent Neural Network ◽  
Fundamental Movement Patterns

Physical inactivity increases the risk of many adverse health conditions, including the world’s major non-communicable diseases, such as coronary heart disease, type 2 diabetes, and breast and colon cancers, shortening life expectancy. There are minimal medical care and personal trainers’ methods to monitor a patient’s actual physical activity types. To improve activity monitoring, we propose an artificial-intelligence-based approach to classify physical movement activity patterns. In more detail, we employ two deep learning (DL) methods, namely a deep feed-forward neural network (DNN) and a deep recurrent neural network (RNN) for this purpose. We evaluate the two models on two physical movement datasets collected from several volunteers who carried tri-axial accelerometer sensors. The first dataset is from the UCI machine learning repository, which contains 14 different activities-of-daily-life (ADL) and is collected from 16 volunteers who carried a single wrist-worn tri-axial accelerometer. The second dataset includes ten other ADLs and is gathered from eight volunteers who placed the sensors on their hips. Our experiment results show that the RNN model provides accurate performance compared to the state-of-the-art methods in classifying the fundamental movement patterns with an overall accuracy of 84.89% and an overall F1-score of 82.56%. The results indicate that our method provides the medical doctors and trainers a promising way to track and understand a patient’s physical activities precisely for better treatment.

scholarly journals Deep Learning for Classifying Physical Activities from Accelerometer Data

2021 ◽  
Author(s):  
Vimala Nunavath ◽  
Sahand Johansen ◽  
Tommy Johannessen ◽  
Lei Jiao ◽  
Bjørge Herman Hansen ◽  
...  
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Activity Patterns ◽  
Physical Activities ◽  
Accelerometer Data ◽  
Feed Forward Neural Network ◽  
Physical Movement ◽  
Colon Cancers ◽  
Deep Recurrent Neural Network ◽  
Fundamental Movement Patterns

Physical inactivity increases the risk of many adverse health conditions, including the world’s major non-communicable diseases, such as coronary heart disease, type 2 diabetes, and breast and colon cancers, shortening life expectancy. There are minimal medical care and personal trainers’ methods to monitor a patient’s actual physical activity types. To improve activity monitoring, we propose an artificial-intelligence-based approach to classify the physical movement activity patterns. In more detail, we employ two deep learning (DL) methods, namely a deep feed-forward neural network (DNN) and a deep recurrent neural network (RNN) for this purpose. We evaluate the proposed models on two physical movement datasets collected from several volunteers who carried tri-axial accelerometer sensors. The first dataset is from the UCI machine learning repository, which contains 14 different activities-of-daily-life (ADL) and is collected from 16 volunteers who carried a single wrist-worn tri-axial accelerometer. The second dataset includes ten other ADLs and is gathered from 8 volunteers who placed the sensors on their hips. Our experiment results show that the RNN model provides the accuracy performance compared to the state-of-the-art methods in classifying the fundamental movement patterns with an overall accuracy of 84.89% and an overall F1-score of 82.56%. Our results indicate that the proposed method will provide the medical doctors and trainers a promising way to precisely track and understand a patient’s physical activities for better treatment.

Peel Learning for Pathway-Related Outcome Prediction

2021 ◽  
Author(s):  
Yuantong Li ◽  
Fei Wang ◽  
Mengying Yan ◽  
Edward Cantu ◽  
Fan Nils Yang ◽  
...  
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Outcome Prediction ◽  
Group Structure ◽  
Penalized Regression ◽  
Small Sample ◽  
Feed Forward Neural Network ◽  
Gene Expressions ◽  
Learning Methods ◽  
Post Surgery

Abstract Motivation Traditional regression models are limited in outcome prediction due to their parametric nature. Current deep learning methods allow for various effects and interactions and have shown improved performance, but they typically need to be trained on a large amount of data to obtain reliable results. Gene expression studies often have small sample sizes but high dimensional correlated predictors so that traditional deep learning methods are not readily applicable. Results In this paper, we proposed peel learning, a novel neural network that incorporates the prior relationship among genes. In each layer of learning, overall structure is peeled into multiple local substructures. Within the substructure, dependency among variables is reduced through linear projections. The overall structure is gradually simplified over layers and weight parameters are optimized through a revised backpropagation. We applied PL to a small lung transplantation study to predict recipients’ post-surgery primary graft dysfunction using donors’ gene expressions within several immunology pathways, where PL showed improved prediction accuracy compared to conventional penalized regression, classification trees, feed-forward neural network, and a neural network assuming prior network structure. Through simulation studies, we also demonstrated the advantage of adding specific structure among predictor variables in neural network, over no or uniform group structure, which is more favorable in smaller studies. The empirical evidence is consistent with our theoretical proof of improved upper bound of PL’s complexity over ordinary neural networks. Availability and Implementation PL algorithm was implemented in Python and the open-source code and instruction will be available at https://github.com/Likelyt/Peel-Learning

scholarly journals A Novel Outdoor Positioning Technique Using LTE Network Fingerprints

Sensors ◽  
2020 ◽  
Vol 20 (6) ◽  
pp. 1691 ◽  
Author(s):  
Da Li ◽  
Yingke Lei ◽  
Haichuan Zhang
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Transfer Learning ◽  
Feed Forward Neural Network ◽  
Research Attention ◽  
Outdoor Environments ◽  
Dispersion Problem ◽  
Fingerprint Database ◽  
The Internet Of Things

In recent years, wireless-based fingerprint positioning has attracted increasing research attention owing to its position-related features and applications in the Internet of Things (IoT). In this paper, by leveraging long-term evolution (LTE) signals, a novel deep-learning-based fingerprint positioning approach is proposed to solve the problem of outdoor positioning. Considering the outstanding performance of deep learning in image classification, LTE signal measurements are converted into location grayscale images to form a fingerprint database. In order to deal with the instability of LTE signals, prevent the gradient dispersion problem, and increase the robustness of the proposed deep neural network (DNN), the following methods are adopted: First, cross-entropy is used as the loss function of the DNN. Second, the learning rate of the proposed DNN is dynamically adjusted. Third, this paper adopted several data enhancement techniques. To find the best positioning fingerprint and method, three types of fingerprint and five positioning models are compared. Finally, by using a deep residual network (Resnet) and transfer learning, a hierarchical structure training method is proposed. The proposed Resnet is used to train with the united fingerprint image database to obtain a positioning model called a coarse localizer. By using the prior knowledge of the pretrained Resnet, feed-forward neural network (FFNN)-based transfer learning is used to train with the united fingerprint database to obtain a better positioning model, called a fine localizer. The experimental results convincingly show that the proposed DNN can automatically learn the location features of LTE signals and achieve satisfactory positioning accuracy in outdoor environments.

scholarly journals HetEnc: A Deep Learning Predictive Model for Multi-type Biological Dataset

2019 ◽  
Author(s):  
Leihong Wu ◽  
Xiangwen Liu ◽  
Joshua Xu
Keyword(s):  
Neural Network ◽  
Gene Expression ◽  
Deep Learning ◽  
Classification Problem ◽  
Biological Data ◽  
Feature Representation ◽  
Biological Research ◽  
Integrated Analysis ◽  
Learning Approaches ◽  
Feed Forward Neural Network

Abstract Background: Researchers today are generating unprecedented amounts of biological data. One trend in current biological research is integrated analysis with multi-platform data. Effective integration of multi-platform data into the solution of a single or multi-task classification problem; however, is critical and challenging. In this study, we proposed HetEnc, a novel deep learning-based approach, for information domain separation. Results: HetEnc includes both an unsupervised feature representation module and a supervised neural network module to handle multi-platform gene expression datasets. It first constructs three different encoding networks to represent the original gene expression data using high-level abstracted features. A six-layer fully-connected feed-forward neural network is then trained using these abstracted features for each targeted endpoint. We applied HetEnc to the SEQC neuroblastoma dataset to demonstrate that it outperforms other machine learning approaches. Although we used multi-platform data in feature abstraction and model training, HetEnc does not need multi-platform data for prediction, enabling a broader application of the trained model by reducing the cost of gene expression profiling for new samples to a single platform. Thus, HetEnc provides a new solution to integrated gene expression analysis, accelerating modern biological research.

An Enhanced Deep Recurrent Neural Network for Autism Spectrum Disorder Diagnosis

2021 ◽  
Vol 11 (12) ◽  
pp. 3028-3037
Author(s):  
D. Pavithra ◽  
A. N. Jayanthi
Keyword(s):  
Neural Network ◽  
Autism Spectrum Disorder ◽  
Deep Learning ◽  
Recurrent Neural Network ◽  
Autism Spectrum ◽  
Spectrum Disorder ◽  
Domain Identification ◽  
Investigation Area ◽  
Deep Recurrent Neural Network ◽  
Sensitivity Specificity

Autism Spectrum Disorder is one of the major investigation area in current era. There are many research works introduced earlier for handling the Autism Spectrum Disorders. However those research works doesn’t achieve the expected accuracy level. The accuracy and prediction efficiency can be increased by building a better classification system using Deep Learning. This paper focuses on the deep learning technique for Autism Diagnosis and the domain identification. In the proposed work, an Enhanced Deep Recurrent Neural Network has been developed for the detection of ASD at all ages. It attempts to predict the autism spectrum in the children along with prediction of areas which can predict the autism in the prior level. The main advantage of EDRNN is to provide higher accuracy in classification and domain identification. Here Artificial Algal Algorithm is used for identifying the most relevant features from the existing feature set. This model was evaluated for the data that followed Indian Scale for Assessment of Autism. The results obtained for the proposed EDRNN has better accuracy, sensitivity, specificity, recall and precision.

Breast Cancer Classification using Discrete Wavelet Transformation and Deep Learning

2019 ◽  
Vol 13 ◽  
Author(s):  
Emmanuel Masa-Ibi ◽  
Rajesh Prasad
Keyword(s):  
Breast Cancer ◽  
Neural Network ◽  
Deep Learning ◽  
Cancer Classification ◽  
Classification Model ◽  
Discrete Wavelet ◽  
Feed Forward Neural Network ◽  
Cancer Data ◽  
Breast Cancer Classification ◽  
The Neural Network

Background: One of the most prevalent sicknesses these days is breast cancer which is common amongst women. This sickness has been in increase to an alarming rate due to the lack of accurate administration of diagnoses. Early and accurate detection is one of the safest ways to cure a breast cancer patient. Objectives: The objective of this study is to proffer a more effective way to accurately classify a cancer sample; whether is Benign or Malignant. Methods: The classification model is based on the data collected from the UCI machine learning repository acquired from Wisconsin hospital called Wisconsin breast cancer data (WBCD). In this study, we preprocessed the dataset using DWT and then test the efficiency of deep learning (DL) for breast cancer classification. The model is developed using a feed-forward neural network and the result is compared with the observed values. Results: The result of the experiment proved the effectiveness of the proposed classification technique. The new technique accomplishes 98.90% accuracy for classifying breast cancer. Conclusions: The result from the experiment shows that the importance of data preprocessing and the efficiency of the neural network over other classification algorithms.

scholarly journals Deep Learning-Based Approach to Fast Power Allocation in SISO SWIPT Systems with a Power-Splitting Scheme

2020 ◽  
Vol 10 (10) ◽  
pp. 3634
Author(s):  
Huynh Thanh Thien ◽  
Pham-Viet Tuan ◽  
Insoo Koo
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
High Performance ◽  
Short Term Memory ◽  
Computation Time ◽  
Power Splitting ◽  
Learning Approaches ◽  
Feed Forward Neural Network ◽  
Transmit Power ◽  
Single Input Single Output

Recently, simultaneous wireless information and power transfer (SWIPT) systems, which can supply efficiently throughput and energy, have emerged as a potential research area in fifth-generation (5G) system. In this paper, we study SWIPT with multi-user, single-input single-output (SISO) system. First, we solve the transmit power optimization problem, which provides the optimal strategy for getting minimum power while satisfying sufficient signal-to-noise ratio (SINR) and harvested energy requirements to ensure receiver circuits work in SWIPT systems where receivers are equipped with a power-splitting structure. Although optimization algorithms are able to achieve relatively high performance, they often entail a significant number of iterations, which raises many issues in computation costs and time for real-time applications. Therefore, we aim at providing a deep learning-based approach, which is a promising solution to address this challenging issue. Deep learning architectures used in this paper include a type of Deep Neural Network (DNN): the Feed-Forward Neural Network (FFNN) and three types of Recurrent Neural Network (RNN): the Layer Recurrent Network (LRN), the Nonlinear AutoRegressive network with eXogenous inputs (NARX), and Long Short-Term Memory (LSTM). Through simulations, we show that the deep learning approaches can approximate a complex optimization algorithm that optimizes transmit power in SWIPT systems with much less computation time.

scholarly journals Evaluation of Different Deep-Learning Models for the Prediction of a Ship’s Propulsion Power

2021 ◽  
Vol 9 (2) ◽  
pp. 116
Author(s):  
Panayiotis Theodoropoulos ◽  
Christos C. Spandonidis ◽  
Nikos Themelis ◽  
Christos Giordamlis ◽  
Spilios Fassois
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Analytical Techniques ◽  
Learning Models ◽  
Feed Forward Neural Network ◽  
Volume Increase ◽  
Adverse Conditions ◽  
Number Of Layers ◽  
The Neural Network ◽  
Propulsion Power

Adverse conditions within specific offshore environments magnify the challenges faced by a vessel’s energy-efficiency optimization in the Industry 4.0 era. As the data rate and volume increase, the analysis of big data using analytical techniques might not be efficient, or might even be infeasible in some cases. The purpose of this study is the development of deep-learning models that can be utilized to predict the propulsion power of a vessel. Two models are discriminated: (1) a feed-forward neural network (FFNN) and (2) a recurrent neural network (RNN). Predictions provided by these models were compared with values measured onboard. Comparisons between the two types of networks were also performed. Emphasis was placed on the different data pre-processing phases, as well as on the optimal configuration decision process for each of the developed deep-learning models. Factors and parameters that played a significant role in the outcome, such as the number of layers in the neural network, were also evaluated.

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

Sensors ◽  
2021 ◽  
Vol 22 (1) ◽  
pp. 174
Author(s):  
Junhyuk Kang ◽  
Jieun Shin ◽  
Jaewon Shin ◽  
Daeho Lee ◽  
Ahyoung Choi
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Input Signal ◽  
Short Term Memory ◽  
Learning Model ◽  
Sensor Data ◽  
Accelerometer Data ◽  
Accelerometer Sensor ◽  
Skeleton Image ◽  
Deep Learning Model

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%.

Sign in / Sign up

Export Citation Format

Share Document