scholarly journals MODELS OF GENERATION OF INPUT DATA OF TRAINING OF NEURAL NETWORK MODULES FOR DIAGNOSTIC OF DISEASES IN UROLOGY

2019 ◽  
pp. 116-122
Author(s):  
Mykola Ivanovych Fedorenko
Keyword(s):  
Neural Network ◽  
Mathematical Model ◽  
Input Data ◽  
Probabilistic Approach ◽  
Training Data ◽  
Digital Data ◽  
Training Methods ◽  
Training Procedure ◽  
Network Modules ◽  
Parameter Values

The subject of the research presented in the article is neural network modules (NNMs), which are used to solve problems in the practice of diagnosing diseases in urology. This work aims to develop a mathematical model for generating a multitude of uroflowmetric parameters, in particular, graphs of uroflowrograms of the required volume, used as input data for NNM training. Objective: to develop a mathematical model for the formation of uroflowmetric parameters using a probabilistic approach based on a uniform "white noise". To develop an effective algorithm for the procedure for generating new parameter values and tools for its implementation. Methods used: NNM training methods, mathematical modeling methods, digital signal processing methods, tools for generating and processing random numerical sequences, digital data filtering methods. The following results were obtained: when creating and implementing a mathematical model for generating a large amount of training data, the requirements of randomness are taken into account when obtaining new values of uroflowmetric parameters. And at the same time, the obtained noise values are filtered to values of a given range, which are percentage-wise comparable to the amplitude value of the uroflowmetric parameter. Conclusions. The scientific novelty of the results is as follows: the NNM training method for recognizing diseases in urology has been improved by developing a mathematical model to generate uroflowmetric parameters for NNM training. The presented model allows you to create the necessary amount of data for training neural network modules in the course of experimental research on the recognition of diseases. The generation of uroflowmetric parameters is based on adding noise to the parameter values. This allows you to change the input data of the NNM training in a given range. This ensures the creation of the required input volume of the NNM training procedure. In the future, this contributes to the testing process of trained neural network modules with reliable information on the diagnosis of diseases in urology.

scholarly journals Robust Approach to Supervised Deep Neural Network Training for Real-Time Object Classification in Cluttered Indoor Environment

2021 ◽  
Vol 11 (15) ◽  
pp. 7148
Author(s):  
Bedada Endale ◽  
Abera Tullu ◽  
Hayoung Shi ◽  
Beom-Soo Kang
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Real Time ◽  
Network Architecture ◽  
Input Data ◽  
Deep Neural Network ◽  
Data Augmentation ◽  
Object Classification ◽  
Training Data ◽  
Gradient Descent Algorithm

Unmanned aerial vehicles (UAVs) are being widely utilized for various missions: in both civilian and military sectors. Many of these missions demand UAVs to acquire artificial intelligence about the environments they are navigating in. This perception can be realized by training a computing machine to classify objects in the environment. One of the well known machine training approaches is supervised deep learning, which enables a machine to classify objects. However, supervised deep learning comes with huge sacrifice in terms of time and computational resources. Collecting big input data, pre-training processes, such as labeling training data, and the need for a high performance computer for training are some of the challenges that supervised deep learning poses. To address these setbacks, this study proposes mission specific input data augmentation techniques and the design of light-weight deep neural network architecture that is capable of real-time object classification. Semi-direct visual odometry (SVO) data of augmented images are used to train the network for object classification. Ten classes of 10,000 different images in each class were used as input data where 80% were for training the network and the remaining 20% were used for network validation. For the optimization of the designed deep neural network, a sequential gradient descent algorithm was implemented. This algorithm has the advantage of handling redundancy in the data more efficiently than other algorithms.

scholarly journals Analysis of COVID-19 Infections on a CT Image Using DeepSense Model

2020 ◽  
Vol 8 ◽  
Author(s):  
Adil Khadidos ◽  
Alaa O. Khadidos ◽  
Srihari Kannan ◽  
Yuvaraj Natarajan ◽  
Sachi Nandan Mohanty ◽  
...  
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Input Data ◽  
Training Data ◽  
Learning Framework ◽  
Learning Classifiers ◽  
Tomography Image ◽  
Mining Model ◽  
Deep Learning Model

In this paper, a data mining model on a hybrid deep learning framework is designed to diagnose the medical conditions of patients infected with the coronavirus disease 2019 (COVID-19) virus. The hybrid deep learning model is designed as a combination of convolutional neural network (CNN) and recurrent neural network (RNN) and named as DeepSense method. It is designed as a series of layers to extract and classify the related features of COVID-19 infections from the lungs. The computerized tomography image is used as an input data, and hence, the classifier is designed to ease the process of classification on learning the multidimensional input data using the Expert Hidden layers. The validation of the model is conducted against the medical image datasets to predict the infections using deep learning classifiers. The results show that the DeepSense classifier offers accuracy in an improved manner than the conventional deep and machine learning classifiers. The proposed method is validated against three different datasets, where the training data are compared with 70%, 80%, and 90% training data. It specifically provides the quality of the diagnostic method adopted for the prediction of COVID-19 infections in a patient.

Segmentation Neural Network Incorporating Scale-Space in the Application of Cardiac MRI

2020 ◽  
Vol 10 (7) ◽  
pp. 1494-1505
Author(s):  
Hyo-Hun Kim ◽  
Byung-Woo Hong
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Cardiac Mri ◽  
Network Architecture ◽  
Data Augmentation ◽  
Scale Space ◽  
Segmentation Algorithm ◽  
Training Data ◽  
Training Procedure ◽  
Whole Heart

In this work, we present an image segmentation algorithm based on the convolutional neural network framework where the scale space theory is incorporated in the course of training procedure. The construction of data augmentation is designed to apply the scale space to the training data in order to effectively deal with the variability of regions of interest in geometry and appearance such as shape and contrast. The proposed data augmentation algorithm via scale space is aimed to improve invariant features with respect to both geometry and appearance by taking into consideration of their diffusion process. We develop a segmentation algorithm based on the convolutional neural network framework where the network architecture consists of encoding and decoding substructures in combination with the data augmentation scheme via the scale space induced by the heat equation. The quantitative analysis using the cardiac MRI dataset indicates that the proposed algorithm achieves better accuracy in the delineation of the left ventricles, which demonstrates the potential of the algorithm in the application of the whole heart segmentation as a compute-aided diagnosis system for the cardiac diseases.

scholarly journals Penerapan Artificial Neural Network Terhadap Identifikasi Wajah Menggunakan Metode Backpropagation

2019 ◽  
Vol 8 (3) ◽  
Author(s):  
Mimin Hendriani ◽  
Rais ◽  
Lilies Handayani
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Network ◽  
Correlation Coefficient ◽  
Training Data ◽  
Face Identification ◽  
Training Methods ◽  
Supervised Training ◽  
R Value ◽  
Artificial Neural

Backpropagation is one of the supervised training methods that causes an error in the output produced. Backpropagation neural networks will be carried out in 3 stages, namely feedforward from input training patterns, backpropagation from errors related to adjustment of weights. Updating the weight is done when the training results obtained have not been converged. The value of the goal error (MSE) is 0.0070579 which is achieved at epochs to 99994 from the provisions of 100000 iterations. Based on the plot regression, the training data resulted in a correlation coefficient value of up to 0.55321. The correlation coefficient value is concluded that the greater the R value produced, the better the level of accuracy in face identification carried out in this study

Development of a Prediction Algorithm for Column Shortening in High-Rise Buildings Using a Neural Network

2007 ◽  
Vol 348-349 ◽  
pp. 901-904
Author(s):  
Won Jik Yang ◽  
Waon Ho Yi
Keyword(s):  
Neural Network ◽  
Training Data ◽  
Prediction Algorithm ◽  
Training Methods ◽  
Training Algorithm ◽  
High Rise ◽  
Concrete Members ◽  
Polynomial Fit ◽  
Neural Network Algorithm ◽  
And Training

The objective of this study is to formulate and evaluate a new training algorithm of Neural Network to predict the inelastic shortening of reinforced concrete members using the column shortening data of high-rise buildings. The new training algorithm of Neural Network for the prediction of column shortening focuses on component of input data and training methods. The validity is examined by training and prediction process based on column shortening measuring data of high-rise buildings. The polynomial fit line of measuring data is used as the training data instead of measuring data. The result shows that the new Neural Network algorithm proposed in this study successfully predicts column shortening of high-rise buildings.

Automatic velocity analysis using convolutional neural network and transfer learning

Geophysics ◽  
2019 ◽  
Vol 85 (1) ◽  
pp. V33-V43 ◽  
Author(s):  
Min Jun Park ◽  
Mauricio D. Sacchi
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Transfer Learning ◽  
Input Data ◽  
Synthetic Data ◽  
Velocity Model ◽  
Training Data ◽  
Velocity Analysis ◽  
Time Step ◽  
Data Set

Velocity analysis can be a time-consuming task when performed manually. Methods have been proposed to automate the process of velocity analysis, which, however, typically requires significant manual effort. We have developed a convolutional neural network (CNN) to estimate stacking velocities directly from the semblance. Our CNN model uses two images as one input data for training. One is an entire semblance (guide image), and the other is a small patch (target image) extracted from the semblance at a specific time step. Labels for each input data set are the root mean square velocities. We generate the training data set using synthetic data. After training the CNN model with synthetic data, we test the trained model with another synthetic data that were not used in the training step. The results indicate that the model can predict a consistent velocity model. We also noticed that when the input data are extremely different from those used for the training, the CNN model will hardly pick the correct velocities. In this case, we adopt transfer learning to update the trained model (base model) with a small portion of the target data to improve the accuracy of the predicted velocity model. A marine data set from the Gulf of Mexico is used for validating our new model. The updated model performed a reasonable velocity analysis in seconds.

A LOCAL-PATCH BASED MULTI-STAGE ARTIFICIAL-NEURAL-NETWORK TRAINING PROCEDURE AND ITS APPLICATION TO MATERIAL CHARACTERIZATION

2007 ◽  
Vol 04 (03) ◽  
pp. 439-458 ◽  
Author(s):  
YUNHUA LUO ◽  
ARVIND SHAH
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Finite Element ◽  
Training Data ◽  
Training Procedure ◽  
Training Time ◽  
Multi Stage ◽  
Training Stage ◽  
Local Patch ◽  
Artificial Neural

A local-patch based multi-stage artificial-neural-network (ANN) training procedure is proposed in this paper, to improve the accuracy of an ANN trained by a backpropagation (BP) algorithm and, at the same time, to reduce the overall training time. In the proposed procedure a conventional one-stage training procedure is split into multiple stages: an initial training stage and subsequent re-training stage(s). In the initial stage the training data are so selected that the trained ANN has adequate ability of generalization, that is, if provided with a set of new input, the ANN can predict the right region where the output is located, but the accuracy of the solution is not necessarily high. In the following re-training stage(s), local patches of training data, either selected from an existing data pool or generated by numerical methods such as finite element method, are used to re-train the ANN to improve the accuracy. Several factors that may have significant effects on the proposed procedure were investigated based on function approximation. As an example of application, the procedure was then used to train an ANN with finite element data to characterize material parameters.

scholarly journals HIGH AVAILABILITY IN SERVER CLUSTERS BY USING BACKPROPAGATION NEURAL NETWORK METHOD

2021 ◽  
Vol 4 (1) ◽  
pp. 08-18
Author(s):  
Ahmad Heryanto ◽  
Aditya Gunanta
Keyword(s):  
Neural Network ◽  
Information Needs ◽  
High Availability ◽  
Training Data ◽  
Backpropagation Neural Network ◽  
Layer 4 ◽  
Device Applications ◽  
Optimal Accuracy ◽  
Hidden Layer ◽  
Parameter Values

Server is a host device applications to serve every request in finding information needs. The server must fully support the services used for the organization's digital needs 24 hours in a day, 7 days in a week, and 365 days in a year. The concept of High Availability is needed to maintain the quality of server services. The algorithm used to build HA can use both classical and modern algorithms. The algorithm used in this research is using backpropagation neural network. In this study, the parameter values to obtain optimal accuracy are learning rate 0.1, training data 80 and test data 20, the number of nodes in hidden layer 4, minimum error 0.0001, and the number of iterations 2500.The best accuracy value using these parameters is 93.79% .

Nonlinear parameter estimation of airship using modular neural network

2019 ◽  
Vol 124 (1273) ◽  
pp. 409-428 ◽  
Author(s):  
S. Agrawal ◽  
D. Gobiha ◽  
N.K. Sinha
Keyword(s):  
Neural Network ◽  
Mathematical Model ◽  
Degrees Of Freedom ◽  
Back Propagation ◽  
Open Loop ◽  
Data Generation ◽  
Modular Neural Network ◽  
Aerodynamic Model ◽  
Parameter Values ◽  
The Mathematical Model

AbstractThe prime focus of this work is to estimate stability and control derivatives of an airship in a completely nonlinear environment. A complete six degrees of freedom airship model has its aerodynamic model as nonlinear functions of angle of attack. Estimating the parameters of aerodynamic model in a nonlinear environment is challenging as it demands an exhaustive dataset that could cover the entire regime of operation of airship. In this work, data generation is achieved by simulating the mathematical model of airship for different trim conditions obtained from continuation analysis. The mathematical model is simulated using predicted parameter values obtained using DATCOM methodology. A modular neural network is then trained using back-propagation and Adam optimisation algorithm for each of the aerodynamic coefficients separately. The estimated nonlinear airship parameters are found to be consistent with the DATCOM parameter values which were used for open-loop simulation. This validates the proposed methodology and could be extended to estimate airship parameters from real flight data.

Sign in / Sign up

Export Citation Format

Share Document