Proposed Convolutional Neural Network Model Research Articles

Enhanced E-commerce Fraud Prediction Based on a Convolutional Neural Network Model

The rapidly escalating sophistication of e-commerce fraud in recent years has led to an increasing reliance on fraud detection methods based on machine learning. However, fraud detection methods based on conventional machine learning approaches suffer from several problems, including an excessively high number of network parameters, which decreases the efficiency and increases the difficulty of training the network, while simultaneously leading to network overfitting. In addition, the sparsity of positive fraud incidents relative to the overwhelming proportion of negative incidents leads to detection failures in trained networks. The present work addresses these issues by proposing a convolutional neural network (CNN) framework for detecting e-commerce fraud, where network training is conducted using historical market transaction data. The number of network parameters reduces via the local perception field and weight sharing inherent in the CNN framework. In addition, this deep learning framework enables the use of an algorithmic-level approach to address dataset imbalance by focusing the CNN model on minority data classes. The proposed CNN model is trained and tested using a large public e-commerce service dataset from 2018, and the test results demonstrate that the model provides higher fraud prediction accuracy than existing state-of-the-art methods.

An Improved Deep Structure for Accurately Brain Tumor Recognition

Brain neoplasms are recognized with a biopsy, which is not commonly done before decisive brain surgery. By using Convolutional Neural Networks (CNNs) and textural features, the process of diagnosing brain tumors by radiologists would be a noninvasive procedure. This paper proposes a features fusion model that can distinguish between no tumor and brain tumor types via a novel deep learning structure. The proposed model extracts Gray Level Co-occurrence Matrix (GLCM) textural features from MRI brain tumor images. Moreover, a deep neural network (DNN) model has been proposed to select the most salient features from the GLCM. Moreover, it manipulates the extraction of the additional high levels of salient features from a proposed CNN model. Finally, a fusion process has been utilized between these two types of features to form the input layer of additional proposed DNN model which is responsible for the recognition process. Two common datasets have been applied and tested, Br35H and FigShare datasets. The first dataset contains binary labels, while the second one splits the brain tumor into four classes; glioma, meningioma, pituitary, and no cancer. Moreover, several performance metrics have been evaluated from both datasets, including, accuracy, sensitivity, specificity, F-score, and training time. Experimental results show that the proposed methodology has achieved superior performance compared with the current state of art studies. The proposed system has achieved about 98.22% accuracy value in the case of the Br35H dataset however, an accuracy of 98.01% has been achieved in the case of the FigShare dataset.

Brain tumor detection using CNN, AlexNet & GoogLeNet ensembling learning approaches

<abstract> <p>The detection of neurological disorders and diseases is aided by automatically identifying brain tumors from brain magnetic resonance imaging (MRI) images. A brain tumor is a potentially fatal disease that affects humans. Convolutional neural networks (CNNs) are the most common and widely used deep learning techniques for brain tumor analysis and classification. In this study, we proposed a deep CNN model for automatically detecting brain tumor cells in MRI brain images. First, we preprocess the 2D brain image MRI image to generate convolutional features. The CNN network is trained on the training dataset using the GoogleNet and AlexNet architecture, and the data model's performance is evaluated on the test data set. The model's performance is measured in terms of accuracy, sensitivity, specificity, and AUC. The algorithm performance matrices of both AlexNet and GoogLeNet are compared, the accuracy of AlexNet is 98.95, GoogLeNet is 99.45 sensitivity of AlexNet is 98.4, and GoogLeNet is 99.75, so from these values, we can infer that the GooGleNet is highly accurate and parameters that GoogLeNet consumes is significantly less; that is, the depth of AlexNet is 8, and it takes 60 million parameters, and the image input size is 227 × 227. Because of its high specificity and speed, the proposed CNN model can be a competent alternative support tool for radiologists in clinical diagnosis.</p> </abstract>

Ethereum Cryptocurrency Entry Point and Trend Prediction using Bitcoin Correlation and Multiple Data Combination

Deep learning methods have achieved significant success in various applications, including trend signal prediction in financial markets. However, most existing approaches only utilize price action data. In this paper, we propose a novel system that incorporates multiple data sources and market correlations to predict the trend signal of Ethereum cryptocurrency. We conduct experiments to investigate the relationship between price action, candlestick patterns, and Ethereum-Bitcoin correlation, aiming to achieve highly accurate trend signal predictions. We evaluate and compare two different training strategies for Convolutional Neural Networks (CNNs), one based on transfer learning and the other on training from scratch. Our proposed 1-Dimensional CNN (1DCNN) model can also identify inflection points in price trends during specific periods through the analysis of statistical indicators. We demonstrate that our model produces more reliable predictions when utilizing multiple data representations. Our experiments show that by combining different types of data, it is possible to accurately identify both inflection points and trend signals with an accuracy of 98%.

A Convolutional Neural Network-Based Maximum Power Point Voltage Forecasting Method for Pavement PV Array

The shadows formed by fast-moving vehicles on a pavement PV array exhibit complex dynamic random distribution characteristics, which can cause a dynamic multipeak PV curve. Dynamic vehicle shadow will cause a reduction in pavement PV power, so the question is how to maximize the power in such conditions by operating at different maximum power point (MPP) quickly and continually. To address this issue, this article proposes an MPP voltage forecasting method based on convolutional neural network (CNN). This method inputs the environmental information of pavement PV array into the proposed CNN model for learning and then uses this model to forecast the MPP voltage. Finally, simulation and experimental test with ResNet, MLP, and CNN methods are carried out and the comparison results show that this model can accurately predict the MPP voltage of pavement PV array under different vehicle shading conditions.

Modulation classification analysis of CNN model for wireless communication systems

<abstract><p>Modulation classification (MC) is a critical task in wireless communication systems, enabling the identification of the modulation class in the received signals. In this paper, we analyzed a novel multi-layer convolutional neural network (CNN) to extract hierarchical features directly from the raw baseband samples. Moreover, we compared the training and testing accuracy of the CNN model for various decimation rates, input sample size and the number of convolutional layers. The results showed that the three-layer CNN model provided better classification accuracy with less computation cost. Furthermore, we observed that the MC performance of the proposed CNN model was better than the other deep learning (DL) and cumulant-based models.</p></abstract>

Deep Transfer Learning Based Detection and Classification of Citrus Plant Diseases

Citrus fruit crops are among the world’s most important agricultural products, but pests and diseases impact their cultivation, resulting in yield and quality losses. Computer vision and machine learning have been widely used to detect and classify plant diseases over the last decade, allowing for early disease detection and improving agricultural production. This paper presented an automatic system for the early detection and classification of citrus plant diseases based on a deep learning (DL) model, which improved accuracy while decreasing computational complexity. The most recent transfer learning-based models were applied to the Citrus Plant Dataset to improve classification accuracy. Using transfer learning, this study successfully proposed a Convolutional Neural Network (CNN)-based pre-trained model (EfficientNetB3, ResNet50, MobiNetV2, and InceptionV3) for the identification and categorization of citrus plant diseases. To evaluate the architecture’s performance, this study discovered that transferring an EfficientNetb3 model resulted in the highest training, validating, and testing accuracies, which were 99.43%, 99.48%, and 99.58%, respectively. In identifying and categorizing citrus plant diseases, the proposed CNN model outperforms other cutting-edge CNN model architectures developed previously in the literature.

Deep Learning-Based Ocular Disease Recognition

This study investigates the use of image classification methods to identify eye disorders using fundus images. Ocular disorders can significantly affect a person's quality of life, but frequent eye exams can help identify them early and prevent vision loss. Manual diagnosis, however, can take a while and is prone to mistakes made by humans. This study suggests utilising deep learning methods to automatically identify eye disorders from fundus photos. To classify several eye disorders, such as age-related macular degeneration, cataracts, and glaucoma, a convolutional neural network (CNN) model is created and trained using a sizable dataset of fundus images. The proposed CNN model achieves high accuracy in the classification of ocular diseases, demonstrating the potential of automated diagnosis for early detection and prevention of vision loss. The results of this research indicate that image classification techniques can significantly improve the accuracy and speed of ocular disease recognition, paving the way for improved ocular health management.

A Novel Approach to Detect Cardiac Arrhythmia Based on Continuous Wavelet Transform and Convolutional Neural Network

Electrocardiogram (ECG) signal is informative as well as non-invasive clinical tool to diagnose cardiac diseases of human heart. However, the diagnosis requires professionals’ clarification and is also time-consuming. To make the diagnosis proficient, a novel convolutional neural network (CNN) has been proposed for automatic arrhythmia detection. In this work, the ECG data collected from the MIT-BIH database have been preprocessed, and segmented in short ECG segments of 60 s. Then, all these segments have been transformed into scalogram images obtained from time-frequency analysis using continuous wavelet transform (CWT). Finally, these scalogram images have been exploited as an input for our designed CNN classifier to classify cardiac arrhythmia. In this approach, the overall accuracy, sensitivity, and specificity are 99.39%, 98.79%, and 100% respectively. Proposed CNN model has significant advantages, and it can be used to differentiate the healthy and arrhythmic patients effectively.

A novel convolutional neural network structure for differential diagnosis of wide QRS complex tachycardia

Background and objectivesCardiac arrhythmias are a significant cause of morbidity and mortality in patients with cardiovascular disease. Accurate rhythm diagnosis is critical in patients presenting with wide QRS complex tachycardia (WCT). Real-time visual interpretation of electrocardiograms (ECG) of complex arrhythmias is difficult and requires expertise. We designed a convolutional neural network (CNN) that could accurately classify WCT into those that are ventricular in origin (ventricular tachycardia (VT)) or supraventricular tachycardia with aberrancy (SVT). MethodsA total of 3065 patients with wide complex ECGs were screened (415 with VT and 2650 with SVT). A CNN model was designed through a Neural Architecture Search (NAS) method. This CNN consisted of a stem convolution layer and five cells, each cell containing separable-convolution and dilated-separable-convolution layers. ResultsUsing 5-fold cross-validation and executing algorithm for five independent runs (with five different seeds), the proposed CNN model achieved a detection accuracy of 87.5 ± 0.0025 and 91.7 %±0.0004 for VT and SVT, respectively. The total sensitivity, specificity, positive predictive value, negative predictive value and F1-score of the CNN model were 88.50 %, 88.50 %, 88.54 %, 88.54 %, and 88.49 %, respectively. ConclusionsIn a cohort of patients presenting with a WCT, our CNN model achieved an accuracy of 87.5% and 91.7% to correctly diagnose VT and SVT, respectively. This model has the potential of being used in real-time settings and to assist physicians with interpretation and decision making.

Convolutional neural network-based high-precision and speed detection system on CIDDS-001

The growing interconnection of complex infrastructures gives very advanced communication functionalities, which gives a massive increase in connected devices and an associated flow volume. Cloud computing is constantly threatened by sophisticated attacks, which poses challenges for a security system. The Cloud will obsolete existing detection procedures against cyber-attacks where they would not be adapted accordingly. Intrusion detection is a classification problem wherein various machine learning and data mining techniques are applied to classify the network data into normal and attack traffic. Therefore, the proposal of new rapid and effective detection approaches is an absolute necessity. In this work, a proposed framework is a network anomaly detection system based on Deep Learning. The analysis carried out was based on the hyper-parameters of the layers of our model. This proposed model is a combination of two techniques; namely, a reduction of dimensions based on the approach of the main components and the second is based on a dense supervised neural network based on convolution neural network (CNN) for a multi-classification of normal and intrusive events from a recent data-set Coburg Network Intrusion Detection data-set (CIDDS-001). The experiments carried out show that the very precise choice of hyper-parameters gives better results. By running the proposed CNN model, it is capable of detecting attacks with an accuracy of 99.13% and an execution time of 12 s.

A multi-view feature fusion approach for effective malware classification using Deep Learning

The number of malware infected machines from all over the world has been growing day by day. New malware variants appear in the wild to evade the malware detection and classification systems and may infect with ransomware or crypto miners for adversary financial gain. A recent colonial pipeline ransomware attack is an example of these attacks that impacted daily human activities, and the victim had to pay ransom to restore their operations. Windows-based systems are the most adopted systems across different industries for running applications. They are prone to get targeted by installing the malware. In this paper, we propose a Deep Learning (DL)-based Convolutional Neural Network (CNN) model to perform the malware classification on Portable Executable (PE) binary files using the fusion feature set approach. We present an extensive performance evaluation of various DL model architecture and Machine Learning (ML) classifier i.e. Support Vector Machine (SVM), on multi-aspect feature sets covering the static, dynamic, and image features to select the proposed CNN model. We further leverage the CNN-based architecture for effective classification of the malware using different combinations of feature sets and compare the results with the best-performed individual feature set. Our performance evaluation of the proposed model shows that the model classifies the malware or benign files with an accuracy of 97% when using fusion feature sets. The proposed model is robust and generalizable and showed similar performances on completely unseen two malware datasets. In addition, the embedding features of the CNN model are visualized, and various visualization methods are employed to understand the characteristics of the datasets. Further, large-scale learning and stacked classifiers were employed after the penultimate layer to enhance the CNN classification performance.

QInward variability-based in-silico proarrhythmic risk assessment of drugs using deep learning model.

Many researchers have suggested evaluation methods and Torsades de Pointes (TdP) metrics to assess the proarrhythmic risk of a drug based on the in silico simulation, as part of the Comprehensive in-vitro Proarrhythmia Assay (CiPA) project. In the previous study, we validated the robustness of 12 in silico features using the ordinal logistic regression (OLR) model by comparing the classification performances of metrics according to the in-vitro experimental datasets used; however, the OLR model using 12 in silico features did not provide desirable results. This study proposed a convolutional neural network (CNN) model using the variability of promising in silico TdP metrics hypothesizing that the variability of in silico features based on beats has more information than the single value of in silico features. We performed the action potential (AP) simulation using a human ventricular myocyte model to calculate seven in silico features representing the electrophysiological cell states of drug effects over 1,000 beats: qNet, qInward, intracellular calcium duration at returning to 50% baseline (CaD50) and 90% baseline (CaD90), AP duration at 50% repolarization (APD50) and 90% repolarization (APD90), and dVm/dtMax_repol. The proposed CNN classifier was trained using 12 train drugs and tested using 16 test drugs among CiPA drugs. The torsadogenic risk of drugs was classified as high, intermediate, and low risks. We determined the CNN classifier by comparing the classification performance according to the variabilities of seven in silico biomarkers computed from the in silico drug simulation using the Chantest dataset. The proposed CNN classifier performed the best when using qInward variability to classify the TdP-risk drugs with 0.94 AUC for high risk and 0.93 AUC for low risk. In addition, the final CNN classifier was validated using the qInward variability obtained after merging three in-vitro datasets, but the model performance decreased to a moderate level of 0.75 and 0.78 AUC. These results suggest the need for the proposed CNN model to be trained and tested using various types of drugs.

CNN Model for Sleep Apnea Detection Based on SpO2 Signal

Sleep Apnea-Hypopnea Syndrome (SAHS) is one of the common sleep disorders which cause hypertension, coronary artery disease, stroke, and diabetes mellitus, as well as the increment of vehicle collisions. Polysomnography is a traditional way of diagnosing sleep disorder which requires multiple sensors for producing multiple physiological signals. Traditional Polysomnography causes huge costs for diagnosing SAHS because it requires numerous sensors as well as time. This study has developed a model by using deep learning techniques to minimize the cost and time for SAHS diagnosing. This study has utilized the SpO2 signal by using a Convolutional Neural Network (CNN) as a deep learning technique to detect SAHS in any individuals. The sleep disorder depends on the amount of blood in the body which is detected by the SpO2 signal.  The proposed CNN model consists of eight layers: three convolution layers, three max-pooling layers, one fully connected layer, and one softmax layer. Two datasets were used: the Apnea-ECG and UCD databases; the first has eight subjects, and the last has 25 subjects. In carrying out the tests, our model achieved an accuracy of 95.5% with the Apnea-ECG database and 90.2% with the UCD database. The suggested technique has provided a cost-effective and efficient way of identifying SAHS in any individual. 

Prediction of Fruit Maturity, Quality, and Its Life Using Deep Learning Algorithms

Fruit that has reached maturity is ready to be harvested. The prediction of fruit maturity and quality is important not only for farmers or the food industry but also for small retail stores and supermarkets where fruits are sold and purchased. Fruit maturity classification is the process by which fruits are classified according to their maturity in their life cycle. Nowadays, deep learning (DL) has been applied in many applications of smart agriculture such as water and soil management, crop planting, crop disease detection, weed removal, crop distribution, strong fruit counting, crop harvesting, and production forecasting. This study aims to find the best deep learning algorithms which can be used for the prediction of fruit maturity and quality for the shelf life of fruit. In this study, two datasets of banana fruit are used, where we create the first dataset, and the second dataset is taken from Kaggle, named Fruit 360. Our dataset contains 2100 images in 3 categories: ripe, unripe, and over-ripe, each of 700 images. An image augmentation technique is used to maximize the dataset size to 18,900. Convolutional neural networks (CNN) and AlexNet techniques are used for building the model for both datasets. The original dataset achieved an accuracy of 98.25% for the CNN model and 81.75% for the AlexNet model, while the augmented dataset achieved an accuracy of 99.36% for the CNN model and 99.44% for the AlexNet model. The Fruit 360 dataset achieved an accuracy of 81.96% for CNN and 81.75% for the AlexNet model. We concluded that for all three datasets of banana images, the proposed CNN model is the best suitable DL algorithm for bananas’ fruit maturity classification and quality detection.

A CNN Based Anomaly Detection Network for Utility Tunnel Fire Protection

Fire accident is one of the significant threats to the urban utility tunnel (UUT) during operation, and the emergency response is challenging due to the compact tunnel structure and potential hazard sources involved. Traditional fire detection techniques are reviewed in this study, and it has been determined that their performance cannot satisfy the requirements for early fire incident detection. Integrating advanced sensing technologies and data-driven anomaly detection has recently been regarded as a feasible solution for intelligent safety system implementation. This article proposed an approach that utilized a fiber-optic distributed temperature sensing (FO-DTS) system and deep anomaly detection models to monitor the fire exotherm during the early stages of accidents. The variable fire exotherm is simulated with an embedded-system controlled electrical heating platform. Moreover, autoencoder (AE) based and convolutional neural network (CNN) based methods have been designed for anomaly detection. The temperature data collected from the FO-DTS in the experiment was employed as the training set for the data-driven models. Furthermore, the anomaly detection models were tested, and the results showed that the proposed CNN model can achieve a higher accuracy rate in detecting the simulated fire exotherm.

Reconstructing turbulent velocity information for arbitrarily gappy flow fields using the deep convolutional neural network

Detailed and reliable flow information is the basis for understanding and further mediating turbulent flows. Due to experimental limitations, such as the absence of seeding particles owing to an inhomogeneous tracer distribution or obstructed optical paths, gappy flow-field data frequently appear with diverse shapes. To resolve this problem, we propose herein the use of a convolutional neural network (CNN) model to reconstruct the velocity field with the missing information of wall-confined turbulent flows. We consider the example of a turbulent channel flow with a frictional Reynolds number Reτ=180 and use machine learning to attain the given objective. High-fidelity numerical data obtained by direct numerical simulation based on the lattice Boltzmann equation are used to generate the datasets required for network training, where data in randomly located square or rectangular regions are masked to provide a maximally realistic instantaneous gappy flow field. The results show that the missing information in gappy regions can be effectively reconstructed for both instantaneous and temporally continuous flow fields. Furthermore, the results are insensitive to the missing locations, even if the locations vary with time. The L2 relative error of the reconstructed instantaneous flow field is generally around 2%. Furthermore, an analysis based on the kinetic-energy spectrum and proper orthogonal decomposition verifies that the reconstructed data are physically consistent with the ground truth. The extracted dominating modes have a maximum relative error level of 10−3. The results obtained herein verify that the proposed CNN model provides complete and reliable data for gappy flows and are physically consistent with physical data.

Spatio-Temporal Image Representation and Deep-Learning-Based Decision Framework for Automated Vehicles

Driving maneuver decision-making is critical to the development and mass deployment of automated vehicles (AVs). The prevailing approaches are stuck with either specific optimization objectives or separate maneuver planning. Inspired by the inherent driving mechanism of human beings, we develop a comprehensive deep learning scheme that extracts and establishes informative image-based representation of dynamic traffic flows and then learns driving strategies from collected driving datasets based on convolutional neural networks (CNNs). Our scheme can effectively capture comprehensive and dynamic features of traffic flows surrounding the ego car and construct dynamic motion images through processing the spatio-temporal signals of all neighbor cars. These constructed virtual images cover all informative dynamic states of the neighbor cars, located in the maneuver-critical area, defined as the motion-sensitive area (MSA). Then, sequential spatio-temporal images together with labeled driving behaviors are fed into our proposed CNN model. The network can extract underlying motion patterns and learn proper driving behaviors, including both the lateral maneuver and longitudinal speed. We demonstrate the effectiveness of the proposed scheme in a typical highway scenario. Results suggest that the proposed scheme merits further investigation to promote the launch of AVs.

Pneumonia detection through Image Classification Using CNN

Pneumonia is a dangerous and serious lung disease. Generally, Chest Radiographs are used to detect pneumonia. This article describes a new method based on Convolutional Neural Network (CNN) classifiers that can identify pneumonia from a patient’s X-ray. Hence, we are proposing pneumonia detection based on CNN using chest X-ray. In this paper, we analyze the existing methods of pneumonia detection and classification techniques. The performance of the proposed CNN model is compared with VGG19 and ResNet50-V2. Performance evaluation is done using Precision and Recall. The proposed CNN model provides higher accuracy compared to existing CNN models.

Improving the Robustness and Quality of Biomedical CNN Models through Adaptive Hyperparameter Tuning

Deep learning is an obvious method for the detection of disease, analyzing medical images and many researchers have looked into it. However, the performance of deep learning algorithms is frequently influenced by hyperparameter selection, the question of which combination of hyperparameters are best emerges. To address this challenge, we proposed a novel algorithm for Adaptive Hyperparameter Tuning (AHT) that automates the selection of optimal hyperparameters for Convolutional Neural Network (CNN) training. All of the optimal hyperparameters for the CNN models were instantaneously selected and allocated using a novel proposed algorithm Adaptive Hyperparameter Tuning (AHT). Using AHT, enables CNN models to be highly autonomous to choose optimal hyperparameters for classifying medical images into various classifications. The CNN model (Deep-Hist) categorizes medical images into basic classes: malignant and benign, with an accuracy of 95.71%. The most dominant CNN models such as ResNet, DenseNet, and MobileNetV2 are all compared to the already proposed CNN model (Deep-Hist). Plausible classification results were obtained using large, publicly available clinical datasets such as BreakHis, BraTS, NIH-Xray and COVID-19 X-ray. Medical practitioners and clinicians can utilize the CNN model to corroborate their first malignant and benign classification assessment. The recommended Adaptive high F1 score and precision, as well as its excellent generalization and accuracy, imply that it might be used to build a pathologist’s aid tool.