A Workflow for the Performance of the Differential Ovarian Follicle Count Using Deep Neuronal Networks

2020 ◽  
pp. 019262332096913
Author(s):  
Eleonora Carboni ◽  
Heike Marxfeld ◽  
Hanati Tuoken ◽  
Christian Klukas ◽  
Till Eggers ◽  
...  
Keyword(s):  
Neural Network ◽  
Performance Metrics ◽  
Ovarian Follicle ◽  
Network Architectures ◽  
The Neural Network ◽  
Good Laboratory ◽  
Different Depths ◽  
Set Up ◽  
Sensitivity Specificity ◽  
And Storage

In order to automate the counting of ovarian follicles required in multigeneration reproductive studies performed in the rat according to Organization for Economic Co-operation and Development guidelines 443 and 416, the application of deep neural networks was tested. The manual evaluation of the differential ovarian follicle count is a tedious and time-consuming task that requires highly trained personnel. In this regard, deep learning outputs provide overlay pictures for a more detailed documentation, together with an increased reproducibility of the counts. To facilitate the planned good laboratory practice (GLP) validation a workflow was set up using MLFlow to make all steps from generating of scans, training of the neural network, uploading of study images to the neural network, generation and storage of the results in a compliant manner controllable and reproducible. PyTorch was used as main framework to build the Faster region-based convolutional neural network for the training. We compared the performances of different depths of ResNet models with specific regard to the sensitivity, specificity, accuracy of the models. In this paper, we describe all steps from data labeling, training of networks, and the performance metrics chosen to evaluate different network architectures. We also make recommendation on steps, which should be taken into consideration when GLP validation is aimed for.

Combined Trajectory Planning and Tracking for Autonomous Vehicles on Deformable Terrains

2020 ◽  
Author(s):  
James Dallas ◽  
Yifan Weng ◽  
Tulga Ersal
Keyword(s):  
Neural Network ◽  
Trajectory Planning ◽  
Autonomous Vehicles ◽  
Performance Metrics ◽  
Soft Soil ◽  
Computation Time ◽  
Operating Conditions ◽  
The Neural Network ◽  
Vehicle Response ◽  
The Impact

Abstract In this work, a novel combined trajectory planner and tracking controller is developed for autonomous vehicles operating on off-road deformable terrains. Common approaches to trajectory planning and tracking often rely on model-dependent schemes, which utilize a simplified model to predict the impact of control inputs to future vehicle response. However, in an off-road context and especially on deformable terrains, accurately modeling the vehicle response for predictive purposes can be challenging due to the complexity of the tire-terrain interaction and limitations of state-of-the-art terramechanics models in terms of operating conditions, computation time, and continuous differentiability. To address this challenge and improve vehicle safety and performance through more accurate prediction of the plant response, in this paper, a nonlinear model predictive control framework is presented that accounts for terrain deformability explicitly using a neural network terramechanics model for deformable terrains. The utility of the proposed scheme is demonstrated on high fidelity simulations for a notional lightweight military vehicle on soft soil. It is shown that the neural network based controller can outperform a baseline Pacejka model based scheme by improving on performance metrics associated with the cost function. In more severe maneuvers, the neural network based controller can achieve sufficient fidelity as compared to the plant to complete maneuvers that lead to failure for the Pacejka based controller. Finally, it is demonstrated that the proposed framework is conducive to real-time implementability.

A Promising Hardware Accelerator with PAST Adder

2021 ◽  
Vol 105 ◽  
pp. 241-248
Author(s):  
Abhishek Choubey ◽  
Shruti Bhargava Choubey
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Resource Utilization ◽  
Network Inference ◽  
Hardware Accelerator ◽  
High Complexity ◽  
High Resource ◽  
The Neural Network ◽  
And Storage ◽  
Dedicated Hardware

Recent neural network research has demonstrated a significant benefit in machine learning compared to conventional algorithms based on handcrafted models and features. In regions such as video, speech and image recognition, the neural network is now widely adopted. But the high complexity of neural network inference in computation and storage poses great differences on its application. These networks are computer-intensive algorithms that currently require the execution of dedicated hardware. In this case, we point out the difficulty of Adders (MOAs) and their high-resource utilization in a CNN implementation of FPGA .to address these challenge a parallel self-time adder is implemented which mainly aims at minimizing the amount of transistors and estimating different factors for PASTA, i.e. field, power, delay.

Empirical mean curve decomposition with multiwavelet transformation for eye movements recognition using electrooculogram signals

2020 ◽  
Vol 234 (8) ◽  
pp. 794-811
Author(s):  
Harikrishna Mulam ◽  
Malini Mudigonda
Keyword(s):  
Neural Network ◽  
Eye Movements ◽  
False Positive Rate ◽  
False Negative ◽  
False Negative Rate ◽  
Principal Component ◽  
The Neural Network ◽  
Marquardt Algorithm ◽  
Positive Rate ◽  
Sensitivity Specificity

Many research works are in progress in classification of the eye movements using the electrooculography signals and employing them to control the human–computer interface systems. This article introduces a new model for recognizing various eye movements using electrooculography signals with the help of empirical mean curve decomposition and multiwavelet transformation. Furthermore, this article also adopts a principal component analysis algorithm to reduce the dimension of electrooculography signals. Accordingly, the dimensionally reduced decomposed signal is provided to the neural network classifier for classifying the electrooculography signals, along with this, the weight of the neural network is fine-tuned with the assistance of the Levenberg–Marquardt algorithm. Finally, the proposed method is compared with the existing methods and it is observed that the proposed methodology gives the better performance in correspondence with accuracy, sensitivity, specificity, precision, false positive rate, false negative rate, negative predictive value, false discovery rate, F1 score, and Mathews correlation coefficient.

Study on the Deformation Forecasting for the Tunnel Surrounding Rock Using the Neural Network

2013 ◽  
Vol 353-356 ◽  
pp. 828-832
Author(s):  
Guo Feng Wang ◽  
Wen Zhao ◽  
Yong Ping Guan ◽  
Lei Liang
Keyword(s):  
Neural Network ◽  
Maximum Error ◽  
Mining Area ◽  
Surrounding Rock ◽  
Rock Deformation ◽  
The Neural Network ◽  
Sublevel Caving ◽  
Iron Mine ◽  
Set Up ◽  
Surrounding Rock Deformation

The non-pillar sublevel caving method is used in Iron Mine in Banshi. In the mining area, there are many folds and faults, the inclination of ore body changes greatly, and ore and rock are fragmentized. The tunnel often collapsed and the surrounding rock deformation was getting large during the construction stage. Using the data of tunnel surrounding rock deformation, we adopt the neural network method to set up the mapping relation between the tunnel surrounding rock deformation and the project factors, including tunnel deepness, tunnel dimension, measuring time and surrounding rock quality. The analyzing results show that the maximum error between the forecast and the testing data is 13%, which indicates that this method is useful and feasible to the mining engineering. Key words: rock pressure; measure, deformation of the tunnel surrounding rock; neural network; data normalization; mapping

scholarly journals Comparison of Feedforward and Recurrent Neural Network in Forecasting Chaotic Dynamical System

2019 ◽  
Vol 10 (37) ◽  
pp. 31-44
Author(s):  
Engin Kandıran ◽  
Avadis Hacınlıyan
Keyword(s):  
Neural Network ◽  
Dynamical System ◽  
Time Series ◽  
Function Approximation ◽  
Feedforward Neural Network ◽  
Network Architectures ◽  
Chaotic Dynamical System ◽  
Global Function ◽  
The Neural Network ◽  
Hidden Layer

Artificial neural networks are commonly accepted as a very successful tool for global function approximation. Because of this reason, they are considered as a good approach to forecasting chaotic time series in many studies. For a given time series, the Lyapunov exponent is a good parameter to characterize the series as chaotic or not. In this study, we use three different neural network architectures to test capabilities of the neural network in forecasting time series generated from different dynamical systems. In addition to forecasting time series, using the feedforward neural network with single hidden layer, Lyapunov exponents of the studied systems are forecasted.

scholarly journals An Automatic Logistics System using Artificial Intelligence

2019 ◽  
Vol 8 (2S4) ◽  
pp. 924-926
Keyword(s):  
Neural Network ◽  
Artificial Intelligence ◽  
Neural Networks ◽  
Artificial Neural Network ◽  
Optimal Route ◽  
Logistics System ◽  
The Neural Network ◽  
Artificial Intelligence Technology ◽  
Artificial Neural ◽  
Set Up

In this paper, we propose a method to utilize machine learning to automate the system of classifying and transporting large quantities of logistics. First, establish an environment similar to the task of transferring logistics to the desired destination, and set up basic rules for classification and transfer. Next, each of the logistics that need sorting and transportation is defined as one entity, and artificial intelligence is introduced so that each individual can go to an optimal route without collision between the objects to the destination. Artificial intelligence technology uses artificial neural networks and uses genetic algorithms to learn neural networks. The artificial neural network is generated by each chromosome, and it is evolved based on the most suitable artificial neural network, and a score is given to each operation to evaluate the fitness of the neural network. In conclusion, the validity of this algorithm is evaluated through the simulation of the implemented system.

scholarly journals Cancer Diagnosis of Microscopic Biopsy Images Using a Social Spider Optimisation-Tuned Neural Network

Diagnostics ◽  
2021 ◽  
Vol 12 (1) ◽  
pp. 11
Author(s):  
Prasanalakshmi Balaji ◽  
Kumarappan Chidambaram
Keyword(s):  
Neural Network ◽  
Performance Metrics ◽  
Disease Diagnosis ◽  
Medical Community ◽  
Diagnostic Tools ◽  
Clinical Cancer Research ◽  
Cancer Disease ◽  
Social Spider ◽  
Life Threatening ◽  
Sensitivity Specificity

One of the most dangerous diseases that threaten people is cancer. If diagnosed in earlier stages, cancer, with its life-threatening consequences, has the possibility of eradication. In addition, accuracy in prediction plays a significant role. Hence, developing a reliable model that contributes much towards the medical community in the early diagnosis of biopsy images with perfect accuracy comes to the forefront. This article aims to develop better predictive models using multivariate data and high-resolution diagnostic tools in clinical cancer research. This paper proposes the social spider optimisation (SSO) algorithm-tuned neural network to classify microscopic biopsy images of cancer. The significance of the proposed model relies on the effective tuning of the weights of the neural network classifier by the SSO algorithm. The performance of the proposed strategy is analysed with performance metrics such as accuracy, sensitivity, specificity, and MCC measures, and the attained results are 95.9181%, 94.2515%, 97.125%, and 97.68%, respectively, which shows the effectiveness of the proposed method for cancer disease diagnosis.

GA-BP Neural Network Based Tire Noise Prediction

2012 ◽  
Vol 443-444 ◽  
pp. 65-70 ◽  
Author(s):  
Yong Che ◽  
Wang Xin Xiao ◽  
Li Jun Chen ◽  
Zhi Chu Huang
Keyword(s):  
Neural Network ◽  
Optimal Solution ◽  
Noise Prediction ◽  
Bp Network ◽  
Tire Noise ◽  
The Neural Network ◽  
Highly Nonlinear ◽  
Neural Network Structure ◽  
Trained Neural Network ◽  
Set Up

According to the complexity and the highly nonlinear characteristics of the tire sound, various parameters affecting tire noise were analyzed. By employing neural network a new method of tire noise prediction was proposed. Combining BP neural networks with genetic algorithms the noise prediction model was set up. In order to effectively predict tire noise, the neural network structure was designed and the input and output parameters of the network were determined. The genetic algorithm was added to the BP network in order to optimize initial weights and search out the optimal solution of the network. Applying laboratory drum method large amounts of tire noise test samples were obtained to train the BP network. Trained neural network can accurately predict tire noise in range of typical frequency bands. The results show that precision of this method is sufficient and the prediction effect is better.

scholarly journals Modelo para detectar el uso correcto de mascarillas en tiempo real utilizando redes neuronales convolucionales

2021 ◽  
Vol 9 (17) ◽  
pp. 111-120
Author(s):  
Hugo Andrade Carrera ◽  
Soraya Sinche Maita ◽  
Pablo Hidalgo Lascano
Keyword(s):  
Neural Network ◽  
Performance Metrics ◽  
Face Mask ◽  
Training Process ◽  
Public Places ◽  
The Neural Network ◽  
Great Performance ◽  
The Family ◽  
The Face ◽  
Mandatory Use

Since Covid-19 appeared, the world has entered into a new stage, in which everybody is trying to mitigate the effects of the virus. The mandatory use of face masks in public places and when maintaining contact with people outside the family circle is one of mandatory measures that many countries have implemented, such as Ecuador, thus, the purpose of this article is to develop a convolutional neural network model using TensorFlow based on MobileNetV2, that allows to perform mask detection in real time video with the key feature of determining if the person is using the face mask properly or if it is not wearing a mask, in order to use the model with OpenCV and a pretrained neural network that detects faces. In addition, the performance metrics of the neural network are analyzed, including precision, accuracy, recall and the F1 score. All performance metrics consider the number of epochs for the training process, obtaining as a result a model that classifies between three groups: faces without face mask, faces wearing a face mask improperly and faces wearing a mask properly. with a great performance in all metrics; The results show values greater than 85% for precision, recall and F1 score, and accuracy values between 93% for 5 epochs and 95% for 25 epochs.

scholarly journals Cinemática Inversa com Redes Neurais Aplicadas em Robôs Manipuladores

2018 ◽  
Vol 1 (43) ◽  
pp. 49
Author(s):  
Gabriel Daltro Duarte ◽  
Claudio Pereira Mego Quinteros ◽  
Lincoln Machado Araújo
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Nonlinear Equations ◽  
Inverse Kinematics ◽  
Robot Manipulator ◽  
Kinematic Model ◽  
Online Training ◽  
Network Architectures ◽  
The Neural Network ◽  
Neural Network Architectures

Today’s world is going through what is known as the Fourth Industrial Revolution. Robots have been gaining more and more space in the industry and going beyond expectations. The use of robots in industry is related to the increasing production and the quality of the electronic products. For an accurate movement of a robot manipulator it is necessary to obtain its inverse kinematic model, however, obtaining this model requires the challenging solution of a set of nonlinear equations. For this system of nonlinear equations there is no generic solution method. In view of this matter, this research aims to solve the problem of the inverse kinematics of a robot manipulator using artificial neural networks without the need to model the robot’s direct kinematics. Two neural network training methodologies, called offline training and online training were used. The basic difference between these two is that in the offline methodology all training points are obtained before any training of the neural network occurs, whereas in online training the use of a training method is recurrent. As the robot moves, new training points are obtained and training processes with the new acquired points are executed, allowing a learning process of continuous inverse kinematics. To validate the proposed methodology a prototype of a manipulated planar robot with one degree of freedom was developed and several architectures of neural networks were tested to find the optimal architecture. The offline training methodology obtained very satisfactory results for most of the neural network architectures tested. The online training only achieved satisfactory results in neural network architectures with quantities of neurons much larger than the quantities used in the architectures used in offline training and still obtained inferior results. The neural networks trained in offline mode, when compared to the training networks in the online mode, presented a greater capacity of generalization and a smaller value of output error. The online training has only achieved satisfactory results in neural network architectures with quantities of neurons much larger than the quantities used in the trained architectures in offline mode.

Sign in / Sign up

Export Citation Format

Share Document