scholarly journals An ensemble-based convolutional neural network model powered by a genetic algorithm for melanoma diagnosis

2021 ◽  
Author(s):  
Eduardo Pérez ◽  
Sebastián Ventura
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Genetic Algorithms ◽  
Convolutional Neural Network ◽  
Network Architecture ◽  
Web Application ◽  
Data Augmentation ◽  
Processing Unit ◽  
Central Processing ◽  
Melanoma Diagnosis

AbstractMelanoma is one of the main causes of cancer-related deaths. The development of new computational methods as an important tool for assisting doctors can lead to early diagnosis and effectively reduce mortality. In this work, we propose a convolutional neural network architecture for melanoma diagnosis inspired by ensemble learning and genetic algorithms. The architecture is designed by a genetic algorithm that finds optimal members of the ensemble. Additionally, the abstract features of all models are merged and, as a result, additional prediction capabilities are obtained. The diagnosis is achieved by combining all individual predictions. In this manner, the training process is implicitly regularized, showing better convergence, mitigating the overfitting of the model, and improving the generalization performance. The aim is to find the models that best contribute to the ensemble. The proposed approach also leverages data augmentation, transfer learning, and a segmentation algorithm. The segmentation can be performed without training and with a central processing unit, thus avoiding a significant amount of computational power, while maintaining its competitive performance. To evaluate the proposal, an extensive experimental study was conducted on sixteen skin image datasets, where state-of-the-art models were significantly outperformed. This study corroborated that genetic algorithms can be employed to effectively find suitable architectures for the diagnosis of melanoma, achieving in overall 11% and 13% better prediction performances compared to the closest model in dermoscopic and non-dermoscopic images, respectively. Finally, the proposal was implemented in a web application in order to assist dermatologists and it can be consulted at http://skinensemble.com.

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 Comparative evaluation of performance and scalability of convolutional neural network implementations on a multisystem HPC architecture

2021 ◽  
Vol 2062 (1) ◽  
pp. 012008
Author(s):  
Sunil Pandey ◽  
Naresh Kumar Nagwani ◽  
Shrish Verma
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
High Performance ◽  
Processing Unit ◽  
Neural Network Training ◽  
Central Processing ◽  
Network Training ◽  
Hot Rolled ◽  
Hot Rolled Steel ◽  
Performance Computing

Abstract The convolutional neural network training algorithm has been implemented for a central processing unit based high performance multisystem architecture machine. The multisystem or the multicomputer is a parallel machine model which is essentially an abstraction of distributed memory parallel machines. In actual practice, this model corresponds to high performance computing clusters. The proposed implementation of the convolutional neural network training algorithm is based on modeling the convolutional neural network as a computational pipeline. The various functions or tasks of the convolutional neural network pipeline have been mapped onto the multiple nodes of a central processing unit based high performance computing cluster for task parallelism. The pipeline implementation provides a first level performance gain through pipeline parallelism. Further performance gains are obtained by distributing the convolutional neural network training onto the different nodes of the compute cluster. The two gains are multiplicative. In this work, the authors have carried out a comparative evaluation of the computational performance and scalability of this pipeline implementation of the convolutional neural network training with a distributed neural network software program which is based on conventional multi-model training and makes use of a centralized server. The dataset considered for this work is the North Eastern University’s hot rolled steel strip surface defects imaging dataset. In both the cases, the convolutional neural networks have been trained to classify the different defects on hot rolled steel strips on the basis of the input image. One hundred images corresponding to each class of defects have been used for the training in order to keep the training times manageable. The hyperparameters of both the convolutional neural networks were kept identical and the programs were run on the same computational cluster to enable fair comparison. Both the convolutional neural network implementations have been observed to train to nearly 80% training accuracy in 200 epochs. In effect, therefore, the comparison is on the time taken to complete the training epochs.

scholarly journals Chaotic Harris Hawks Optimization with Quasi-Reflection-Based Learning: An Application to Enhance CNN Design

Sensors ◽  
2021 ◽  
Vol 21 (19) ◽  
pp. 6654
Author(s):  
Jameer Basha ◽  
Nebojsa Bacanin ◽  
Nikola Vukobrat ◽  
Miodrag Zivkovic ◽  
K. Venkatachalam ◽  
...  
Keyword(s):  
Neural Network ◽  
Brain Tumor ◽  
Convolutional Neural Network ◽  
Swarm Intelligence ◽  
Network Architecture ◽  
Data Augmentation ◽  
Superior Performance ◽  
Basic Algorithm ◽  
Experimental Conditions ◽  
Neural Network Approach

The research presented in this manuscript proposes a novel Harris Hawks optimization algorithm with practical application for evolving convolutional neural network architecture to classify various grades of brain tumor using magnetic resonance imaging. The proposed improved Harris Hawks optimization method, which belongs to the group of swarm intelligence metaheuristics, further improves the exploration and exploitation abilities of the basic algorithm by incorporating a chaotic population initialization and local search, along with a replacement strategy based on the quasi-reflection-based learning procedure. The proposed method was first evaluated on 10 recent CEC2019 benchmarks and the achieved results are compared with the ones generated by the basic algorithm, as well as with results of other state-of-the-art approaches that were tested under the same experimental conditions. In subsequent empirical research, the proposed method was adapted and applied for a practical challenge of convolutional neural network design. The evolved network structures were validated against two datasets that contain images of a healthy brain and brain with tumors. The first dataset comprises well-known IXI and cancer imagining archive images, while the second dataset consists of axial T1-weighted brain tumor images, as proposed in one recently published study in the Q1 journal. After performing data augmentation, the first dataset encompasses 8.000 healthy and 8.000 brain tumor images with grades I, II, III, and IV and the second dataset includes 4.908 images with Glioma, Meningioma, and Pituitary, with 1.636 images belonging to each tumor class. The swarm intelligence-driven convolutional neural network approach was evaluated and compared to other, similar methods and achieved a superior performance. The obtained accuracy was over 95% in all conducted experiments. Based on the established results, it is reasonable to conclude that the proposed approach could be used to develop networks that can assist doctors in diagnostics and help in the early detection of brain tumors.

scholarly journals Comparative analysis of augmented datasets performances of age invariant face recognition models

2021 ◽  
Vol 10 (3) ◽  
pp. 1356-1367
Author(s):  
Kennedy Okokpujie ◽  
Etinosa Noma-Osaghae ◽  
Samuel Ndueso John ◽  
Charles Ndujiuba ◽  
Imhade Princess Okokpujie
Keyword(s):  
Neural Network ◽  
Face Recognition ◽  
Convolutional Neural Network ◽  
Network Architecture ◽  
Data Augmentation ◽  
Absolute Error ◽  
Extraction Process ◽  
Deep Convolutional Neural Network ◽  
Non Invasive ◽  
Invasive Method

The popularity of face recognition systems has increased due to their non-invasive method of image acquisition, thus boasting the widespread applications. Face ageing is one major factor that influences the performance of face recognition algorithms. In this study, the authors present a comparative study of the two most accepted and experimented face ageing datasets (FG-Net and morph II). These datasets were used to simulate age invariant face recognition (AIFR) models. Four types of noises were added to the two face ageing datasets at the preprocessing stage. The addition of noise at the preprocessing stage served as a data augmentation technique that increased the number of sample images available for deep convolutional neural network (DCNN) experimentation, improved the proposed AIFR model and the trait aging features extraction process. The proposed AIFR models are developed with the pre-trained Inception-ResNet-v2 deep convolutional neural network architecture. On testing and comparing the models, the results revealed that FG-Net is more efficient over Morph with an accuracy of 0.15%, loss function of 71%, mean square error (MSE) of 39% and mean absolute error (MAE) of -0.63%.

scholarly journals On the issue of fuzzy timing estimations of the algorithms running at GPU and CPU architectures

2019 ◽  
Vol 135 ◽  
pp. 01082 ◽  
Author(s):  
Oleg Agibalov ◽  
Nikolay Ventsov
Keyword(s):  
Genetic Algorithm ◽  
Genetic Algorithms ◽  
Graphics Processing Unit ◽  
Processing Unit ◽  
Running Time ◽  
Average Value ◽  
Central Processing ◽  
Graphics Processing ◽  
Number Of Individuals ◽  
Gpu Architecture

We consider the task of comparing fuzzy estimates of the execution parameters of genetic algorithms implemented at GPU (graphics processing unit’ GPU) and CPU (central processing unit) architectures. Fuzzy estimates are calculated based on the averaged dependencies of the genetic algorithms running time at GPU and CPU architectures from the number of individuals in the populations processed by the algorithm. The analysis of the averaged dependences of the genetic algorithms running time at GPU and CPU-architectures showed that it is possible to process 10’000 chromosomes at GPU-architecture or 5’000 chromosomes at CPUarchitecture by genetic algorithm in approximately 2’500 ms. The following is correct for the cases under consideration: “Genetic algorithms (GA) are performed in approximately 2, 500 ms (on average), ” and a sections of fuzzy sets, with a = 0.5, correspond to the intervals [2, 000.2399] for GA performed at the GPU-architecture, and [1, 400.1799] for GA performed at the CPU-architecture. Thereby, it can be said that in this case, the actual execution time of the algorithm at the GPU architecture deviates in a lesser extent from the average value than at the CPU.

scholarly journals Verte-Box: A Novel Convolutional Neural Network for Fully Automatic Segmentation of Vertebrae in CT Image

Tomography ◽  
2022 ◽  
Vol 8 (1) ◽  
pp. 45-58
Author(s):  
Bing Li ◽  
Chuang Liu ◽  
Shaoyong Wu ◽  
Guangqing Li
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Automatic Segmentation ◽  
Feature Representation ◽  
Dice Similarity Coefficient ◽  
Processing Unit ◽  
Central Processing ◽  
Multi Scale ◽  
Vertebrae Segmentation ◽  
Ct Volume

Due to the complex shape of the vertebrae and the background containing a lot of interference information, it is difficult to accurately segment the vertebrae from the computed tomography (CT) volume by manual segmentation. This paper proposes a convolutional neural network for vertebrae segmentation, named Verte-Box. Firstly, in order to enhance feature representation and suppress interference information, this paper places a robust attention mechanism on the central processing unit, including a channel attention module and a dual attention module. The channel attention module is used to explore and emphasize the interdependence between channel graphs of low-level features. The dual attention module is used to enhance features along the location and channel dimensions. Secondly, we design a multi-scale convolution block to the network, which can make full use of different combinations of receptive field sizes and significantly improve the network’s perception of the shape and size of the vertebrae. In addition, we connect the rough segmentation prediction maps generated by each feature in the feature box to generate the final fine prediction result. Therefore, the deep supervision network can effectively capture vertebrae information. We evaluated our method on the publicly available dataset of the CSI 2014 Vertebral Segmentation Challenge and achieved a mean Dice similarity coefficient of 92.18 ± 0.45%, an intersection over union of 87.29 ± 0.58%, and a 95% Hausdorff distance of 7.7107 ± 0.5958, outperforming other algorithms.

Non-Blind Image Deconvolution Based on “Ringing” Removal Using Convolutional Neural Network

2020 ◽  
Vol 2020 (10) ◽  
pp. 181-1-181-7
Author(s):  
Takahiro Kudo ◽  
Takanori Fujisawa ◽  
Takuro Yamaguchi ◽  
Masaaki Ikehara
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Network Architecture ◽  
Large Scale ◽  
Blind Deconvolution ◽  
Training Dataset ◽  
Image Deconvolution ◽  
Classic Problem ◽  
Key Points ◽  
Blind Image

Image deconvolution has been an important issue recently. It has two kinds of approaches: non-blind and blind. Non-blind deconvolution is a classic problem of image deblurring, which assumes that the PSF is known and does not change universally in space. Recently, Convolutional Neural Network (CNN) has been used for non-blind deconvolution. Though CNNs can deal with complex changes for unknown images, some CNN-based conventional methods can only handle small PSFs and does not consider the use of large PSFs in the real world. In this paper we propose a non-blind deconvolution framework based on a CNN that can remove large scale ringing in a deblurred image. Our method has three key points. The first is that our network architecture is able to preserve both large and small features in the image. The second is that the training dataset is created to preserve the details. The third is that we extend the images to minimize the effects of large ringing on the image borders. In our experiments, we used three kinds of large PSFs and were able to observe high-precision results from our method both quantitatively and qualitatively.

Sign in / Sign up

Export Citation Format

Share Document