Convolutional Neural Network Technique Research Articles

Knee Osteoarthritis SCAENet: Adaptive Knee Osteoarthritis Severity Assessment Using Spatial Separable Convolution with Attention-Based Ensemble Networks with Hybrid Optimization Strategy.

Osteoarthritis (OA) of the knee is a chronic state that significantly lowers the quality of life for its patients. Early detection and lifetime monitoring of the progression of OA are necessary for preventive therapy. In the course of therapy, the Kellgren and Lawrence (KL) assessment model categorizes the rigidity of OA. Deep techniques have recently been used to increase the precision and effectiveness of OA severity assessments. The training process is compromised by low-confidence samples, which are less accurate than normal ones. In this work, a deep learning-based knee osteoarthritis severity assessment model is recommended to accurately identify the condition in patients. The phases of the designed model are data collection, feature extraction, and prediction. At first, the images are generally gathered from online resources. The gathered images are given into the feature extraction phase. A new model is implemented to predict knee osteoarthritis named Spatial Separable Convolution with Attention-based Ensemble Networks (SCAENet), which includes feature extraction, stacked target-based feature pool generation, and knee osteoarthritis prediction. The feature extraction is done using ResNet, Visual Geometry Group (VGG16), and DenseNet. The stacked target-based feature pool is obtained from the SCAENet. Hence, the stacked target-based feature pool is obtained by the Hybridization of Equilibrium Slime Mould with Bald Eagle Search Optimization (HESM-BESO). Here, the knee osteoarthritis's severity prediction is performed using the dimensional convolutional neural network (1DCNN) technique. The designed SCAENet model is validated with other conventional methods to show high performance.

Quantum classical hybrid convolutional neural networks for breast cancer diagnosis.

The World Health Organization states that early diagnosis is essential to increasing the cure rate for breast cancer, which poses a danger to women's health worldwide. However, the efficacy and cost limitations of conventional diagnostic techniques increase the possibility of misdiagnosis. In this work, we present a quantum hybrid classical convolutional neural network (QCCNN) based breast cancer diagnosis approach with the goal of utilizing quantum computing's high-dimensional data processing power and parallelism to increase diagnosis efficiency and accuracy. When working with large-scale and complicated datasets, classical convolutional neural network (CNN) and other machine learning techniques generally demand a large amount of computational resources and time. Their restricted capacity for generalization makes it challenging to maintain consistent performance across multiple data sets. To address these issues, this paper adds a quantum convolutional layer to the classical convolutional neural network to take advantage of quantum computing to improve learning efficiency and processing speed. Simulation experiments on three breast cancer datasets, GBSG, SEER and WDBC, validate the robustness and generalization of QCCNN and significantly outperform CNN and logistic regression models in classification accuracy. This study not only provides a novel method for breast cancer diagnosis but also achieves a breakthrough in breast cancer diagnosis and promotes the development of medical diagnostic technology.

Time-evolving traffic resilience performance forecasting during hazardous weather toward proactive intervention

Transportation systems experience significant disruptions and loss during hazardous weather events, exhibiting great needs of timely intervention to effectively improve the resilience of the affected traffic systems. An informed and science-based proactive intervention strategy depends on accurate forecasting of the resilience performance of traffic systems with essential lead time during hazards. A new resilience performance forecasting methodology at both global and local scales is proposed for traffic networks under natural hazards by addressing unique challenges such as scarcity and time-evolving nature of hazard-specific data. The proposed methodology consists of two modules: the local traffic resilience performance short-term forecasting module based on the modified diffusion convolutional recurrent neural network (DCRNN) and transfer learning techniques, and the global traffic resilience performance forecasting module integrating percolation-based robustness assessment and SIR-based congestion propagation modeling. A case study of an urban traffic network during a major snowstorm hazard is conducted as a demonstration, followed by the feasibility investigation to guide proactive intervention during hazards. It is found the proposed methodology can forecast the time-evolving traffic resilience performance with good accuracy at both global and local scales. With sufficient lead time for the forecast, it bears promising potential to assist the stakeholders to make informed and timely decision about possible proactive intervention by providing key information to help identify the optimal moments and individual strategic links for possible intervention.

Enhanced Malaysian License Plate Recognition System Using an Improved YOLOv2 Model

License Plate Recognition (LPR) has gained popularity among researchers due to its wide range of applications, including law enforcement, monitoring, and toll gate systems. However, existing LPR systems still require improvements to achieve optimum accuracy and speed. The advancements in Convolutional Neural Network (CNN) variants offer potential solutions for these challenges. This primary aim of this system is to ensure accurate and efficient recognition of the vehicle plate characters using CNN techniques. This research utilizes two CNN network architectures for deep object detection to address the Malaysian License Plate Recognition (MLPR) task. The first network is designed to detect the license plate, while the second is responsible for recognizing the characters on the plate. Both networks are cascaded from the architecture of two-stage YOLOv2, providing promising speed and accuracy. The MLPR system achieved an accuracy of 98.75% and a processing speed of 0.0104 seconds, using a total of 2,200 license plate images. In conclusion, the system adapted from deep object detection techniques presents a promising solution for the MLPR problem, based on the achieved accuracy and speed.

Convolutional neural network based detection of early stage Parkinson’s disease using the six minute walk test

The heterogeneity of Parkinson’s disease (PD) presents considerable challenges for accurate diagnosis, particularly during early-stage disease, when the symptoms may be extremely subtle. This study aimed to assess the accuracy of a convolutional neural network (CNN) technique based on the 6-min walk test (6MWT) measured using wearable sensors to distinguish patients with early-stage PD (n = 78) from healthy controls (n = 50). The participants wore six sensors, and performed the 6MWT. The time-series data were converted into new images. The results revealed that the gyroscopic vertical component of the lumbar spine displayed the highest classification accuracy of 83.5%, followed by those of the thoracic spine (83.1%) and right thigh (79.5%) segment. These findings suggest that the 6MWT and CNN models may facilitate earlier diagnosis and monitoring of PD symptoms, enabling clinicians to provide timely treatment during the critical transition from normal to pathologic gait patterns.

Lung cancer classification model using convolutional neural network with feature ranking process

Abstract Lung cancer is the leading cause of cancer-related deaths worldwide, highlighting the importance of early detection to improve patient outcomes. The goal of this study is to create a computer-aided diagnosis (CAD) system that detects and classifies lung cancer based on medical images using a Convolutional Neural Network (CNN) and feature extraction techniques. The study employs the LIDC-IDRI dataset, a comprehensive collection of lung cancer-related medical images, to achieve this goal. Contrast Limited Adaptive Histogram Equalization (CLAHE) is used to improve contrast and detail while reducing noise. The Sobel operator is used to refine the segmentation process by enhancing edges and contours in binary images. Morphological operations such as dilation and filling are used to refine the segmented regions even further. Feature extraction is used to extract statistical data and texture characteristics from segmented regions. Mean and variance calculations reveal information about brightness and variability within regions, whereas co-occurrence matrices and Gray-Level Co-occurrence Matrix (GLCM) properties quantify texture features. The correlation between different regions is also evaluated to assess their relationships. Using the pre-processed and ranked features as inputs, a CNN model with five hidden layers is trained. To classify the segmented regions into cancerous and non-cancerous classes, the model learns patterns and relationships in the data. A confusion matrix is used to assess the accuracy, specificity, and sensitivity of the model's predictions, with an emphasis on correctly identifying lung cancer-affected regions. The results show promising results, with the proposed CAD system identifying lung cancer-affected regions with an accuracy of 99.4375%. The system also outperforms other existing methods with a specificity of 99.12% and a sensitivity of 99.26%. These findings highlight the developed system's potential as a valuable tool for early lung cancer detection, assisting doctors in making accurate diagnoses and improving patient outcomes.

Predicting long-term deposit customers using convolutional neural network and data conversion technique

Predicting long-term deposit customers using convolutional neural network and data conversion technique

Detection of bone cancer based on a four-phase framework generative deep belief neural network in deep learning

The medical image processing plays an important role in analyzing and detecting tumors and detecting cancer cells in their early stages. Microscopic images were used as a deep learning approach to recognize bone features. These images were acquired using microscopy, a time-consuming and labor-intensive process that required repeating. Cancer cells cannot be identified by conventional techniques because of the noise in the images. To address these challenges, this paper introduces the Gated Convolutional Neural Network (GCNN) techniques used to identify multiclass bone cancer, including non-tumor, non-viable, and viable tumors. As a first stage, the most basic part of image Preprocessing is to deposing without interrupting the diagnostics information during the removal of noise. The second stage is edge detection, used for image brightness discontinuity detection based on Scaled Region Edge Detector (SRED). It is used for image segmentation and data extraction within regions based on the Screen Cluster Area Segmentation (SCAS) to segment the Region of Interest (RoI) based on Relative Scalar Color Histogram (RSCH). The fourth step is Feature extraction is to find out the features and match the image feature from the Bone Cancer Detection object based on the maximum threshold weights using GCNN with Maximum Standard External Regions (MSER) used for features extracting applied to improve feature quality analysis and Feature support rate. This results in higher classification accuracy, Precision, recall under F-measure, and lower error rates. This enables a more achieved classification than other existing methods.

Employee attrition prediction with convolutional neural network and synthetic minority over-sampling technique

ABSTRACT Employee attrition negatively affects organizational efficiency, customer service quality, and brand reputation. In this study, we explore the important issue of predicting employee attrition, which aids managers in taking effective retention strategies and mitigating the significant costs associated with recruiting and training new personnel. Different from existing research, we introduce a novel deep learning model featuring an intermediary layer that automatically generates a hidden image representation from tabular data. This intermediary step facilitates the efficient utilization of Convolutional Neural Networks, specifically designed for image data, thereby enhancing predictive accuracy. Furthermore, we employ the widely used Synthetic Minority Over-sampling Technique for handling imbalanced data to further improve our model’s performance. Our new CNN-based deep learning model demonstrated the best performance for predicting employee attrition, which can assist in reducing costs related to turnover and facilitate the implementation of a succession plan to ensure seamless transitions.

A Novel Transformer-CNN Approach for Predicting Soil Properties from LUCAS Vis-NIR Spectral Data

Soil, a non-renewable resource, requires continuous monitoring to prevent degradation and support sustainable agriculture. Visible-near-infrared (Vis-NIR) spectroscopy is a rapid and cost-effective method for predicting soil properties. While traditional machine learning methods are commonly used for modeling Vis-NIR spectral data, large datasets may benefit more from advanced deep learning techniques. In this study, based on the large soil spectral library LUCAS, we aimed to enhance regression model performance in soil property estimation by combining Transformer and convolutional neural network (CNN) techniques to predict 11 soil properties (clay, silt, pH in CaCl2, pH in H2O, CEC, OC, CaCO3, N, P, and K). The Transformer-CNN model accurately predicted most soil properties, outperforming other methods (partial least squares regression (PLSR), random forest regression (RFR), support vector machine regression (SVR), Long Short-Term Memory (LSTM), ResNet18) with a 10–24 percentage point improvement in the coefficient of determination (R2). The Transformer-CNN model excelled in predicting pH in CaCl2, pH in H2O, OC, CaCO3, and N (R2 = 0.94–0.96, RPD > 3) and performed well for clay, sand, CEC, P, and K (R2 = 0.77–0.85, 2 < RPD < 3). This study demonstrates the potential of Transformer-CNN in enhancing soil property prediction, although future work should aim to optimize computational efficiency and explore a wider range of applications to ensure its utility in different agricultural settings.

Identifying liver cancer cells using cascaded convolutional neural network and gray level co-occurrence matrix techniques

<p>Liver cancer has a high mortality rate, especially in South Asia, East Asia, and Sub-Saharan Africa. Efforts to reduce these rates focus on detecting liver cancer at all stages. Early detection allows more treatment options, though symptoms may not always be apparent. The staging process evaluates tumor size, location, lymph node involvement, and spread to other organs. Our research used the CLD staging system, assessing tumor size (C), lymph nodes (L), and distant invasion (D). We applied a deep learning approach with a cascaded convolutional neural network (CNN) and gray level co-occurrence matrix (GLCM)-based texture features to distinguish benign from malignant tumors. The method validated with the cancer imaging archive (TCIA) dataset, demonstrating superior accuracy compared to existing techniques.</p>

Deciphering the Complexities of COVID‐19‐Related Cardiac Complications: Enhancing Classification Accuracy With an Advanced Deep Learning Framework

ABSTRACTThe literature has widely described the interaction between cardiac complications and COVID‐19. However, the diagnosis of cardiac complications caused by COVID‐19 using Computed Tomography (CT) images remains a challenge due to the diverse clinical manifestations. To address this issue, this study proposes a novel configuration of Convolutional Neural Network (CNN) for detecting cardiac complications derived from COVID‐19 using CT images. The main contribution of this work lies in the use of CNN techniques in combination with Long Short‐Term Memory (LSTM) for cardiac complication detection. To explore two‐class classification (COVID‐19 without cardiac complications vs. COVID‐19 with cardiac complications), 10 650 CT images were collected from COVID‐19 patients with and without myocardial infarction, myocarditis, and arrhythmia. The information was annotated by two radiology specialists. A total of 0.926 was found to be the accuracy, 0.84 was the recall, 0.82 was the precision, 0.82 was the F1‐score, and 0.830 was the g‐mean of the suggested model. These results show that the suggested approach can identify cardiac problems from COVID‐19 in CT scans. Patients with COVID‐19 may benefit from the proposed LSTM‐CNN architecture's enhanced ability to identify cardiac problems.

Apple fruits categorizing based on deep convolutional neural network techniques

<p>For a variety of reasons, including the high degree of similarity between varieties of the same type of fruit, the requirement to train the technique on a large amount of data, and the type and number of features suitable for application, the use of computer vision techniques in the classification of fruits still faces many challenges. Additionally, the technique's effectiveness and speed both need to be improved. Deep conventional neural network (DCNN) approaches were required for all of these reasons. A proposed CNN model is described in this work. The suggested methodology is intended to quickly and accurately categorize thirteen groups of apple fruits. The proposed technique was based on training and testing the model on a maximum number of images of apple fruits, by increasing the number of database images tenfold, after augmentation was performed on the images. The technology also relied on good tuning of the hyperparameters. To further ensure the efficiency of training, validation was performed on 20% of the database. All results that demonstrate the high efficiency of the proposed model were reviewed. The results of the proposal were compared with the results of four related techniques. The results showed the great advantage of the proposed technology at all levels.</p>

Bridging spatio-temporal discontinuities in global soil moisture mapping by coupling physics in deep learning

The launch of Soil Moisture Active Passive (SMAP) satellite in 2015 has resulted in significant achievements in global soil moisture mapping. Nonetheless, spatiotemporal discontinuities in the soil moisture products have arisen due to the limitations of its orbit scanning gap and retrieval algorithms. To address these issues, this paper presents a physics-constrained gap-filling method, PhyFill for short. The PhyFill method employs a partial convolutional neural network technique to explore spatial domain features of the original SMAP soil moisture data. Then, it incorporates variations in soil moisture induced by precipitation events and dry-down events as penalty terms in the loss function, thereby accounting for monotonicity and boundary constraints in the physical processes governing the dynamic fluctuations of soil moisture. The PhyFill model was applied to SMAP soil moisture data, resulting in continuous spatially daily soil moisture data on a global scale. Three validation strategies are employed: visual inspection through global pattern, simulated missing-region validation, and soil moisture validation with in situ measurements. The results indicated that the reconstructed soil moisture achieved a higher spatial coverage with satisfactory spatial continuity with neighbouring pixels. The simulated validation of the missing regions revealed that the averaged unbiased root mean square difference (ubRMSD) and correlation coefficient (R) were 0.01 m3/m3 and 0.99, respectively versus the gap filled SMAP product. The core validation sites demonstrated that the reconstructed soil moisture data has a consistent ubRMSD compared with the original SMAP soil moisture data (0.04 m3/m3vs. 0.04 m3/m3). The PhyFill method can generate globally continuous, high accurate soil moisture estimates, providing remarkable support for advanced hydrological applications, e.g., global soil moisture dry-down events and patterns.

A recommendation attack detection approach integrating CNN with Bagging

Due to their open architecture, collaborative filtering recommender systems are susceptible to recommendation attacks, in which attackers inject fake rating data into the system to affect the accuracy of recommendation results. To detect these attacks, numerous detection methods have been designed and proven effective. However, in recent years, deep learning-based recommendation attack models such as GSA-GAN have shown higher concealment, posing new challenges to existing detection methods. Motivated by the need for improved detection, in this paper we propose a new approach called CNN-BAG, which integrates convolutional neural network (CNN) and Bagging (BAG) techniques. CNN-BAG can enhance the detection performance by simultaneously leveraging the deep learning capabilities of CNN and the ensemble learning strengths of Bagging. Firstly, we construct a deep neural network based on CNN as the base learner to automatically extract and learn features of recommendation attacks. Secondly, we use the Bagging algorithm to perform bootstrap sampling on the training data to generate multiple diverse training subsets. The above constructed base learners are then trained on these generated training subsets to produce multiple base classifiers for classifying recommendation attacks. Finally, we combine the base classifiers’ outputs using a majority voting method to obtain the final detection results. To assess the performance of CNN-BAG in detecting recommendation attacks, we compared it against several well-established detection methods on the Movielens-10M and Amazon datasets. Our experiments revealed that CNN-BAG is adept at identifying various attack types, including the deep learning-based recommendation attack models.

Developing Bluetooth phonocardiogram for detecting heart murmurs using hybrid MFCC and LSTM

Cardiovascular disease is a leading global cause of mortality. Most stethoscopes still necessitate the use of tubing, which entails direct physical contact between the healthcare provider and patient. The stethoscope can serve as a means of transmission if it is utilized on individuals who have been diagnosed with airborne and droplet-borne infectious illnesses. A prototype was created to capture heart sounds using a Phonocardiography (PCG) device over website-based Bluetooth connectivity. This approach offers the benefits of being cost-effective, facilitating computer-aided diagnostics, and being wearable. In addition, the primary significance of this study resides in the identification of heart sound irregularities caused by cardio dynamic abnormalities of the heart valves, known as murmurs. The heart sound categorization process utilizes a machine learning model that involves extracting 25 Mel frequency cepstral coefficients (MFCC) as features. The model employs a hybrid approach combining convolutional neural network and long short-term memory (CNN-LSTM) techniques. The research findings indicate that the suggested model achieves an average accuracy rate of 95.9% over five distinct categories, i.e., normal, atrial stenosis, mitral regurgitation, mitral stenosis, and mitral valves prolapse. Further study can be conducted on hardware development by incorporating an infrared sensor at the fingertip of the stethoscope.

A two-stage approach for aircraft detection with convolutional neural network

Over the past few years, object detection has experienced remarkable advancements, primarily attributable to significant progress in deep learning architectures. Nonetheless, the task of identifying aircraft targets within remote sensing images remains a challenging and actively explored area. Presently, there are two main approaches employed for this task: one utilizing convolutional neural network (CNN) techniques and the other relying on conventional methods. In this work, a CNN based architecture is proposed to recognize aircraft types using remote sensing images. The experiments performed on multi-type aircraft remote sensing images (MTARSI) dataset show that the proposed architecture achieves 97.07%, 94.81%, and 94.44% accuracy rates for training, validation and testing sets. The results approve that, the architecture outperforms state of the art models.

A Hybrid Deep Learning Model to Estimate the Future Electricity Demand of Sustainable Cities

Rapid population growth, economic growth, and technological developments in recent years have led to a significant increase in electricity consumption. Therefore, the estimation of electrical energy demand is crucial for the planning of electricity generation and consumption in cities. This study proposes a hybrid deep learning model that combines convolutional neural network (CNN) and long short-term memory (LSTM) techniques, both of which are deep learning techniques, to estimate electrical load demand. A hybrid deep learning model and LSTM model were applied to a dataset containing hourly electricity consumption and meteorological information of a city in Türkiye from 2017 to 2021. The results were evaluated using mean absolute percent error (MAPE), root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R2) metrics. The proposed CNN-LSTM hybrid model was compared to the LSTM model, with lower MAPE, MAE, and RMSE values. Furthermore, the CNN-LSTM model exhibited superior prediction performance with an R2 value of 0.8599 compared to the LSTM model with an R2 value of 0.8086. These results demonstrate the effectiveness of the proposed deep learning model in accurately estimating future electrical load demand to plan electricity generation for sustainable cities.

A FRAMEWORK FOR MANAGEMENT OF LEAKS AND EQUIPMENT FAILURE IN OIL WELLS

Oil is a precious and critical natural energy resource that is used in numerous ways to drive various industries worldwide. The extraction of oil from underground reservoirs is a complex process that requires a lot of planning, careful execution, and risk management. In this paper, CNN is employed to extract relevant features from sensor primary data collected from various wells. Detecting undesirable events such as leaks and equipment failure in oil wells is crucial for preventing safety hazards, environmental damage and financial losses, making it challenging to identify issues in a timely and accurate manner. This dissertation describes a hybrid model for detecting undesirable events in oil and gas wells using a combination of Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) techniques. The CNN architecture enables effective information extraction by applying convolutional layers and pooling operations to identify patterns and spatial dependencies in the data. The extracted features are then fed into an LSTM network, which can capture temporal dependencies and learning long-term patterns. By utilizing LSTM, the model can effectively analyse the time series data and detect the occurrence of undesirable events, such as abnormal pressure, fluid leakage, or equipment malfunction, in oil and gas wells. The hybrid model leveraging CNN for feature extraction and LSTM for detecting undesirable events in the oil and gas industry presents a comprehensive approach to enhance well monitoring and prevent potential hazards. Achieving high accuracy rates of 99.8% for training and 99.78% for testing demonstrates the efficacy of the proposed model in accurately identifying and classifying undesirable events in oil and gas wells.

FLOOD SUSCEPTIBILITY MAPPING USING MACHINE LEARNING IN KENING RIVER, SUB WATERSHED OF BENGAWAN SOLO, TUBAN

Floods are natural occurrences with the potential to cause damage to ecosystems and pose significant threats to human life, resulting in the destruction of property, infrastructure, and socioeconomic challenges. In recent times, flooding in the Sub-Watershed of Bengawan Solo has been linked to the overflowing Kening River in Tuban County. Aim: This study aims to produce a flood susceptibility map to mitigate the frequency of flood occurrences as well as facilitate effective planning for flood disaster risk management. Methodology and results: Flood data is collected from 2016 to 2023 through field surveys, Sentinel-1 satellite imagery, and data from the Development Planning Agency, Tuban County. Integrating remote sensing data from satellite imagery (PlanetScope, Sentinel-2), geographic information systems (GIS), and spatial modeling techniques, a flood susceptibility map is developed for the Kening River catchment. The occurrence of floods in the Kening River area is associated with various factors (11 variables) assessed through the frequency ratio approach, including profil curvature, LS factor, aspect, rainfall, river distance, road distance, building density 100 m, road density 100 m, vegetation type, normalized difference water index (NDWI), and soil adjusted vegetation index (SAVI). The results show flood susceptibility maps utilizing frequency ratio (FR) and convolutional neural network (CNN) techniques. The flood susceptibility map obtained through the CNN method demonstrates a notably high AUC value. The model development generated a validation AUC value of 0.857 for training and 0.856 for testing. Conclusion, significance and impact study: This research provides an understanding into the factors that influence the occurrence of floods in the Kening River catchment area. It also emphasizes the benefit of advanced machine learning approaches in mapping the susceptibility of floods. Furthermore, this study has the potential to be helpful in guiding regional policy decisions and result in enhanced flood risk management measures in Tuban County.