Bulletin of Electrical Engineering and Informatics
Not a member yet
2885 research outputs found
Sort by
Automated 3D convolutional neural network architecture design using genetic algorithm for pulmonary nodule classification
Get PDFCancer of the lungs is considered one of the primary causes of death among patients globally. Early detection contributes significantly to the success of pulmonary cancer treatment. To aid the pulmonary nodule classification, many models for the analysis of medical image utilizing deep learning have been developed. Convolutional neural network (CNN) recently, has attained remarkable results in various image classification tasks. Nevertheless, the CNNs performance is heavily dependent on their architectures which still heavily reliant on human domain knowledge. This study introduces a cutting-edge approach that leverages genetic algorithms (GAs) to automatically design 3D CNN architectures for differentiation between benign and malignant pulmonary nodules. The suggested algorithm utilizes the dataset of lung nodule analysis 2016 (LUNA16) for evaluation. Notably, our approach achieved exceptional model accuracy, with evaluations on the testing dataset yielding up to 95.977%. Furthermore, the algorithm exhibited high sensitivity, showcasing its robust performance in distinguishing between benign and malignant nodules. Our findings demonstrate the outstanding capabilities of the proposed algorithm and show an outstanding performance and attain a state of art solution in lung nodule classification
Speech emotion recognition with optimized multi-feature stack using deep convolutional neural networks
Get PDFThe human emotion in communication plays a significant role that can influence how the context of the message is perceived by others. Speech emotion recognition (SER) is one of a field study that is very intriguing to explore because human-computer interaction (HCI) related technologies such as virtual assistant that are implemented nowadays rarely considered the emotion contained in the information relayed by human speech. One of the most widely used ways to perform SER is by extracting features of speech such as mel frequency cepstral coefficient (MFCC), mel-spectrogram, spectral contrast, tonnetz, and chromagram from the signal and using a one-dimensional (1D) convolutional neural network (CNN) as a classifier. This study shows the impact of implementing a combination of an optimized multi-feature stack and optimized 1D deep CNN model. The result of the model proposed in this study has an accuracy of 90.10% for classifying 8 different emotions performed on the ryerson audio-visual database of emotional speech and song (RAVDESS) dataset
Development and evaluation of a network intrusion detection system for DDoS attack detection using machine learning
Get PDFDistributed denial of service (DDoS) attacks involves disrupting a target system by flooding it with an immense volume of traffic originating from numerous sources. These attacks can disrupt online services, causing financial losses and reputational damage to various organizations. To combat this threat, the proposed network intrusion detection system (NIDS) utilizes machine learning (ML) algorithms trained on the KDDCup99 dataset. This dataset encompasses a diverse array of network traffic patterns, bounded by both regular traffic and various attack types. By training the NIDS on this dataset, it becomes capable of accurately identifying DDoS attacks based on their distinctive patterns. The NIDS model is constructed using ML approaches like random forest (RF), support vector machines (SVM), and naive Bayes (NB). The developed NIDS is evaluated using performance metrics such as accuracy, precision, recall, F1-score, and receiver operating characteristic (ROC) curve. The proposed method demonstrates the NIDS’s accuracy of about 93%, precision of 99% and recall of 92% in detecting DDoS attacks, transforming it into a valuable tool for network security in comparison with the current methods. The study contributes to the domain of network security by providing an effective NIDS solution for detecting the DDoS attacks in the wireless sensor network
Microstrip antenna with reflector and air gap for short range communication in 900 MHz band
Get PDFThis paper proposes a microstrip antenna that was made of a microstrip fed slot with a complimentary stub on a single dielectric medium. This antenna was integrated with a reflector and air gap for the application of short range communication (SRC) in a 900 MHz band. Analyses were made on the dimension of the reflector and the height of the air gap towards the antenna performance. Besides, an antenna field test was done for the propagation distance of the proposed antenna. As a result, with the antenna size of 13,770 mm2 , the measured return loss was -10.79 dB and the directivity gain was 7.44 dBi. Besides, with the effective isotropic radiated power (EIRP) of 7.44 dBm, it was predicted that at 100 m, the received signal would be around 60 to 70 dBm. Therefore, a high gain was produced by using a reflector with air gap and a compact size was achieved if compared to conventional high gain antenna designs such as Yagi Uda. Thus, it is suitable for a communication device such as the SRC application
Integration of MQTT-SN and CoAP protocol for enhanced data communications and resource management in WSNs
Get PDFLightweight communication protocols for wireless sensor networks (WSNs) are unfolding for machine to machine (M2M) communications and thus there is always going to be a possible conflict of interest on which protocol is best suited for any particular application. The two protocols of interest in this study are the message queue telemetry transport protocol for sensor network (MQTT-SN), a variant of message queue telemetry transport (MQTT) protocol and the constrained application protocol (CoAP). There have been studies that reveal that these protocols perform differently based on the underlying network conditions. CoAP experience lower delays than MQTT for higher packet loss and higher delays for lower packet loss. MQTT default communication via a broker is easier to scale compared to CoAP direct request-response paradigm. Although this is a huge advantage over CoAP, it presents the single point-of-failure problem. In this paper we propose an integration of MQTT-CoAP protocol using an abstraction layer that enables both MQTT-SN and CoAP protocol to be used in the same sensor node. Resources are managed by directly modifying sensor node configuration using CoAP protocol. Performance evaluation of these protocols under the integrated scenario shows acceptable levels of latency and energy consumption for internet of thing (IoT) operations
Trajectory tracking control based on genetic algorithm and proportional integral derivative controller for two-wheel mobile robot
Get PDFThis paper uses the genetic algorithm (GA) to optimize the proportional integral derivative (PID) controller parameters to present the motion control design for a two-wheeled mobile robot autonomous system. The GA algorithm determines a collision-free travel curve for a robot with a tangential velocity restriction constraint. A trajectory-tracking controller based on the PID control structure is developed to monitor the calculated route curves for the mobile robot. Simulation results show the effectiveness of the GA-PID controller compared to the PID controller. The GA-PID controller demonstrates improved performance in trajectory tracking and collision avoidance, making it suitable for controlling the motion of two-wheeled mobile robots. The GA's optimization process allows for better tuning of the PID controller parameters, resulting in more efficient and accurate robot motion control. The results suggest that the proposed GA-PID controller is a promising approach for enhancing mobile robots' autonomous navigation capabilities
Based on deep convolutional neural network, COVID-19 identification utilizing computed tomography scans
Get PDFIn the year 2019 specifically, on March 11th, the coronavirus illness two thousand nineteen (COVID-19) was announced a worldwide epidemic due to its rapid spread and lack of treatment options. As a result, infected individuals must be identified and quarantined quickly to prevent the illness from spreading. The method used to test for COVID-19 is called real-time-polymerase chain reaction (RT-PCR), which has problems with having low sensitivity and taking an extended amount of time. Because chest computed tomography (CT) scans are more sensitive than RT-PCR, it follows that such scans can be employed for diagnostic purposes. This study developed a deep convolutional neural network (CNN) approach to detect COVID-19 using CT scan images. An architecture of deep learning (DL) called convolutional neural network computed tomography scans (CT-CNN) was utilized to efficiently identify COVID-19. The findings of our suggested model are highly encouraging, with an accuracy of 96.14%, an F1 score of 96.21%, and a recall of 97.53% when it comes to classifying CT scans as either infected or not infected by COVID-19
Secure Euclidean random distribution for patients’ magnetic resonance imaging privacy protection
Get PDFPatients’ information and images transfer among medical institutes represent a major tool for delivering better healthcare services. However, privacy and security for healthcare information are big challenges in telemedicine. Evidently, even a small change in patients’ information might lead to wrong diagnosis. This paper suggests a new model for hiding patient information inside magnetic resonance imaging (MRI) cover image based on Euclidean distribution. Both least signification bit (LSB) and most signification bit (MSB) techniques are implemented for the physical hiding. A new method is proposed with a very high level of security information based on distributing the secret text in a random way on the cover image. Experimentally, the proposed method has high peak signal to noise ratio (PSNR), structural similarity index metric (SSIM) and reduced mean square error (MSE). Finally, the obtained results are compared with approaches in the last five years and found to be better by increasing the security for patient information for telemedicine
Bio-engineered strategies for osteochondral defect repair
Get PDFDue to the absence of blood vessels and nerves, the regenerative potential of articular cartilage is significantly constrained. This implies that the impact of a ruptured cartilage extends to the entire joint. Osteoarthritis, a health condition, may arise due to injury and the progressive breakdown of joint tissues. The progression of osteoarthritis can be accelerated by the extensive degradation of articular cartilage, which is ranked as the third most prevalent musculoskeletal disorder necessitating rehabilitation, following low back pain and fractures. The existing therapeutic interventions for cartilage repair exhibit limited efficacy and seldom achieve complete restoration of both functional capacity and tissue homeostasis. Emerging technological advancements in the field of tissue engineering hold significant promise for the development of viable substitutes for cartilage tissue, capable of exhibiting functional properties. The overarching strategy involves ensuring that the cell source is enriched with bioactive molecules that facilitate cellular differentiation and/or maturation. This review provides a comprehensive summary of recent advancements in the field of cartilage tissue engineering. Additionally, it offers an overview of recent clinical trials that have been conducted to examine the latest research developments and clinical applications pertaining to weakened articular cartilage and osteoarthritis
Design and implementation of energy-efficient hybrid data aggregation in heterogeneous wireless sensor network
Get PDFHeterogeneous wireless sensor network (HWSN) is a trending technology in both the industrial and academic sectors, consisting of a large number of interconnected sensors. However, higher energy consumption and delay are significant drawbacks of this technology in applications such as military, healthcare, and industrial automation. The main objective of this research is to enhance the energy efficiency of HWSN using a clustering technique. In this article, a novel approach, namely power optimization and hybrid data aggregation (POHDA), is proposed to address these challenges in HWSN. POHDA-HWSN focuses on power optimization and congestion avoidance through effective CH selection using hybrid data aggregation based on parameters such as residual energy, distance, mobility, threshold value of the node, and latency. By weight-based effective cluster head (CH) selection, the energy consumption, end-to-end delay, and overhead during communication are reduced in this network. The POHDA-HWSN approach considers specific parameters to compare the results and outcomes with earlier research such as HCCS-WSN, FMCA-WSN, and APCC-WSN. The results prove that the proposed POHDA-HWSN approach achieves higher energy efficiency and delivery ratio