DSpace Arşivi :: by Yazar "Al-Jumaili, Saif" değerine göre listeleniyor

Yazar "Al-Jumaili, Saif" seçeneğine göre listele

Listeleniyor 1 - 14 / 14

Classification of brain tumors using MRI images based on convolutional neural network and supervised machine learning algorithms
(Institute of Electrical and Electronics Engineers Inc., 2022) Ali, Rawaa; Al-Jumaili, Saif; Duru, Adil Deniz; Uçan, Osman Nuri; Boyacı, Aytuğ; Duru, Dilek Göksel
Brain tumor is abnormal cells that originate from cranial tissue and is considered one of the most destructive diseases, and lead to the cause of death, where the early diagnosis is crucial for accelerating the therapy of brain tumors. Examining the patient's MRI scans is one traditional way of distinguishing brain cancers. The conventional approaches take a long time and are prone to human error, especially when dealing with huge amounts of data and diverse brain tumor classes. Artificial Intelligence (AI) is extremely useful for the strict detection and classification of several diseases in the brain. Convolutional Neural Network (CNN) is one of the modes techniques which act as a tumor classifier due to it shows high effectiveness for diagnosing brain tumors. That's why, in this research, we presented a hybrid method that merged a group of pre-Trained deep learning CNN patterns with a group of supervised classifiers in machine learning called, k-Nearest Neighbor (KNN), Support Vector Machine (SVM), and Linear Discriminant Analysis (LDA). We used an MRI image that consist of images of four brain tumor classes, namely glioma, meningioma, pituitary, and no tumor. We deduced the features extracted from the images by hiring three types of CNN called (GoogleNet, Shuffle-Net, and NasNet-Mobile). Depending upon the experimental consequences, ShuffleN et with SVM achieved the highest results according to the four categories of metrics evaluation that are Accuracy of 98.40%, Precision of 97%, Recall of 96.75%, and Fl-Score of 96.75%. Finally, we compared our results with different state-of-The-Art papers recently published and our proposed method show outperforms compared them.
Classification of covid-19 omicron variant using hybrid deep transfer learning based on x-ray chest images
(Altınbaş Üniversitesi, 2022) Al-Jumaili, Saif
In 2019, the first case of COVID-19 was announced in China in Wuhan Province. Which led to the panic of the world and the declaration of a state of extreme emergency by the World Health Organization. Given that the world was in a state of crisis and closure, the use of deep learning technology provides speed and accuracy in diagnosing disease through chest images. Therefore, in this study, the dental X-Ray images of people infected with the omicron strain of Covid-19 virus were classified in comparison with a group of healthy people. In this study, we used 4 types of pre-trained deep learning algorithms in two ways, the first is using cross-validation and the second is the hybrid method by extracting the features from the models and then applying them to two types of deep learning algorithms (SVM and KNN). Accuracy results were obtained in the first scenario with a percentage of 94%, while in the second scenario, the accuracy results in the SVM classifier are higher than KNN with a difference of 5%, which is 92%. We also compared studies that used X-Ray images to classify COVID-19, as our results showed a clear superiority compared to other studies.
Classification of epileptic seizure features from scalp electrical measurements using KNN and SVM based on fourier transform
(American Institute of Physics Inc., 2022) Al-Azzawi, Athar Hussein Ali; Al-Jumaili, Saif; Ibrahim, Abdullahi Abdu; Duru, Adil Deniz
Epilepsy classification techniques are one of the areas that are still under searching till now as long as there is no specific method for detection seizures. The brain consists of more than 100 billion nerves that generate electrical activity. These activities are recorded using an Electroencephalogram (EEG) by electrodes attached to the scalp. EEG is considered a big footstep in the medical and technical field where it allows the detection of brain disorders. However, this paper aims to identify the most efficient classification algorithm for classifying EEG signals of epileptic seizures. Therefore, we applied two classification techniques namely Support Vector Machine (SVM) and k-Nearest Neighbors (KNN), which rely on the features extracted from the data by the Fast Fourier Transform (FFT) method. The results show SVM obtained the highest accuracy value compared to KNN, accurate scores were 99.5% and 99%, respectively.
Classification of the level of alzheimer’s disease using anatomical magnetic resonance images based on a novel deep learning structure
(CRC Press, 2023) Al-Jumaili, Saif; Al-Azzawi, Athar; Uçan, Osman Nuri; Duru, Adil Deniz
Alzheimer’s is an incurable neurodegenerative disease that generally begins slowly and progresses gradually with time. In the early stage, symptoms of memory loss are mild, while in the late stage it clearly shows the deterioration in cognitive functions. Due to its irreversible nature, early detection reflects positively on reducing restraining the spread and preventing damage to the brain cells thus avoiding reaching the dementia stage. Till now, deep learning is considered to be one of the most significant methodologies used to detect and classify different types of neurological diseases from MRI images. However, in this study, we proposed a novel two-dimensional deep convolutional neural network to classify four stages of Alzheimer’s disease. The dataset consists of four types, namely nondemented, very mild demented, mild demented, and moderate demented subject MR images. First, we have applied a preprocessing technique to resize the image for compliance with our models. Then, we performed Reduce Atmospheric Haze techniques that can decrease the atmospheric haze making all images sharp and clear to feed to the model. We implemented the model 30 times and obtained more than 99.46% for evaluation metrics. The proposed method shows an outstanding performance compared to other papers reported in the literature.
Deep transfer learning methods for classification colorectal cancer based on histology images
(Institute of Electrical and Electronics Engineers Inc., 2022) Alhanaf, Ahmed Sami; Al-Jumaili, Saif; Bilgin, Gökhan; Duru, Adil Deniz; Alyassri, Salam; Balık, Hasan Hüseyin
Deep transfer learning is one of the common techniques used to classify different types of cancer. The goal of this research is to focus on and adopt a fast, accurate, suitable, and reliable for classification of colorectal cancer. Digital histology images are adjustable to the application of convolutional neural networks (CNNs) for analysis and classification, due to the sheer size of pixel data present in them. Which can provide a lot of information about colorectal cancer. We used ten different types of pr-trained models with two type method of classification techniques namely (normal classification and k-fold crosse validation) to classify the tumor tissue, we used two different kinds of datasets were these datasets consisting of three classes (normal, low tumors, and high tumors). Among all these eight models of deep transfer learning, the highest accuracy achieved was 96.6% with Darknet53 for 5-Fold and for normal classification the highest results obtained was 98.7% for ResNet50. Moreover, we compared our result with many other papers in stat-of-the-art, the results obtained show clearly the proposed method was outperformed the other papers.
Diagnosis of Epileptic seizures and Hypoxic-ischemic encephalopathy using Artificial Intelligence based on EEG signal: A review
(Institute of Electrical and Electronics Engineers Inc., 2024) Kadhim, Ezzaddin; Al-Jumaili, Saif; Uçan, Osman Nuri
The brain is the nucleus for cognition and controls voluntary and involuntary activities inside the human body. Any neurological illness, regardless of its cause, will impair the brain's functionality. Certain neurological illnesses manifest symptoms as seizures. Epilepsy and Hypoxic-ischemic Encephalopathy (HIE) are the most similar disorders in symptoms, but at the neurological level, they are two completely different disorders. This difference is measured at the level of neural activity, as Electroencephalography (EEG) is one of the most distinctive tools used to measure neural activity in the brain. Experts use EEG to diagnose disorders through recorded brain activity, including seizures, but the diagnosis process consumes much time and effort. Adopting Artificial Intelligence (AI) techniques to extract the patterns of brain illnesses is a more efficient process for diagnosing disorders because it depends on computing and, thus, has high accuracy in diagnosing brain illnesses. In this research, we reviewed the most effective stages and methods adopted by researchers to diagnose brain disorders based on EEG and artificial intelligence techniques.
Evaluation of deep transfer learning methodologies on the COVID-19 radiographic chest images
(International Information and Engineering Technology Association, 2023) Al-Azzawi, Athar; Al-Jumaili, Saif; Duru, Adil Deniz; Duru, Dilek Göksel; Uçan, Osman Nuri
In 2019, the world had been attacked with a severe situation by the new version of the SARSCOV- 2 virus, which is later called COVID-19. One can use artificial intelligence techniques to reduce time consumption and find safe solutions that have the ability to handle huge amounts of data. However, in this article, we investigated the classification performance of eight deep transfer learning methodologies involved (GoogleNet, AlexNet, VGG16, MobileNet-V2, ResNet50, DenseNet201, ResNet18, and Xception). For this purpose, we applied two types of radiographs (X-ray and CT scan) datasets with two different classes: non-COVID and COVID-19. The models are assessed by using seven types of evaluation metrics, including accuracy, sensitivity, specificity, negative predictive value (NPV), F1- score, and Matthew's correlation coefficient (MCC). The accuracy achieved by the X-ray was 99.3%, and the evaluation metrics that were measured above were (98.8%, 99.6%, 99.6%, 99.0%, 99.2%, and 98.5%), respectively. Meanwhile, the CT scan model classified the images without error. Our results showed a remarkable achievement compared with the most recent papers published in the literature. To conclude, throughout this study, it has been shown that the perfect classification of the radiographic lung images affected by COVID- 19.
Investigation of epileptic seizure signatures classification in EEG using supervised machine learning algorithms
(International Information and Engineering Technology Association, 2023) Al-Jumaili, Saif; Duru, Adil Deniz; Ibrahim, Abdullahi Abdu; Uçan, Osman Nuri
Epilepsy is one of the earnest neurological disorders that require further social attention. Based on the International League Against Epilepsy (ILAE), which classifies the epilepsy term as a number of several seizures that occur in the brain. Electroencephalography (EEG) is considered our brain window to the electrical activity. It is a significant device used for diagnosing multiple brain disorders such as Epilepsy. Moreover, this study used data from Temple University Hospital Seizure Corpus (TUH), which represents an accurate description of the clinical cases for five types of epileptic seizures. Initially, to extract information from EEG signals, three types of feature extraction have been used namely Fast Fourier Transform, Entropy, and Approximate Entropy. Due to the high degree of variance of EEG signals, we implemented a band-pass filter to divide the signals into sub-bands called delta rhythm (0.1 - 4Hz), theta rhythm (5 -9Hz), alpha rhythm (10 - 14Hz), beta rhythm (15- 31Hz), and gamma rhythm (32-100). The feature extraction outcome underwent normalization techniques and was used as input for the classifiers. Support Vector Machine (SVM), Decision Tree (DT), Naive Bayes (NB), and K-Nearest Neighbor (KNN) classifier have implemented in order to classify (1) second epoch length window. In the first scenario, we applied the FFT features to the classifiers, the results showed that SVM obtained the highest value compared to the other classifiers with 96% accuracy, whereas KNN was 92% and the DT and NB were 76% and 67%, respectively. The second scenario was applying entropy features to the classifiers, the results of classification were 91% for SVM and 88% for KNN, while the DT and NB were 76% and 67%, respectively. The last scenario was ApEn, which shows that SVM still gains the highest value, which was 83%, and 76% for KNN, where the DT and NB were 65% and 69%, respectively. From the aforementioned results, we deduced that SVM achieved the best accuracy when applied with the three feature extractions.
Preparation of ZnO thin film for gas sensors using spray pyrolysis technique
(Institute of Electrical and Electronics Engineers Inc., 2023) Abdulhadi, Mithal O.; Al-Jumaili, Saif; Uçan, Osman Nuri
In this article, the focus was on the properties of ZnO thin films and how they relate to the performance of sensors used to find oxygen gas and present an easy technique for producing zinc oxide (ZnO) thin films deposited onto normal glass substrates. The synthesis process includes using a precursor solution with a concentration of 0.1 M, which is carried out by applying a low-cost homemade spray pyrolysis technique. X-ray diffraction data is used to do an expanded structural analysis. The scanning electron microscopy image and chosen region electron diffraction pattern assist the structural investigation. After annealing at 500°C, the crystallite sizes are less than 50 nm. This means that the films can be used in nanomaterial applications like gas sensor devices. The observed pattern formation in XRD indicates the preferred orientation is the (101) plane. Spray pyrolysis often yields superior results for many types of applications, especially when compared to other methods. Using atomic force microscopy revealed a dense and uniform arrangement of nanocrystalline structures. The film is also uniformly distributed throughout the substrate. UV-visible spectroscopy examines the optical characteristics, providing more evidence that nanostructures have formed. These results support the preparation of these nanostructures for use in gas sensing applications. The effect of ZnO was more evident after annealing the film, as shown by the observed changes in energy gap behavior and grain size, which as a result influenced the sensor features. The sensitivity and response time enhancement can be observed while annealing the film at a temperature of 500°C.
Pseudopapilledema diagnosis based on a hybrid approach using deep transfer learning
(Institute of Electrical and Electronics Engineers Inc., 2023) Al-Azzawi, Athar; Al-Jumaili, Saif; Duru, Adil Deniz; Bayat, Oğuz; Kurnaz, Sefer; Uçan, Osman Nuri
This Papilledema is edema caused by elevated pressure inside the brain near the area that leads the optic nerve to reach the eye. If left untreated, this condition can cause severe difficulties, for instance, aberrant optical changes, reduced sharpness of vision, and irreversible blindness. At present, an approach based on image processing for determining the degree of papilledema from color fundus images was given utilizing transfer learning approaches. The used dataset here contains 295 papilledema images, 295 pseudopapilledema images, and 779 control images. For the image preparation, a segmentation optimizer was utilized. The performance of the transfer learning techniques GoogleNet, MobileNetV2, ResNet-18, and ResNet-50 was then compared. Furthermore, Sensitivity and specificity and constructed ROC curves were calculated. The ResNet-50 employing the optimizer ADAM method performed best in the testing, with 98% total accuracy. The findings of the studies demonstrated that a combination of segmentation, optimization models, and transfer learning techniques may be utilized to determine the severity of papilledema automatically. The total accuracy was higher when compared to other similar studies described in the literature.
Recent advances on convolutional architectures in medical applications: classical or quantum
(Institute of Electrical and Electronics Engineers Inc., 2022) Al-Jumaili, Saif; Al-Jumaili, Ahmed; Alyassri, Salam; Duru, Adil Deniz; Uçan, Osman Nuri
Deep learning is one of the most significant advances in AI (AI). It is used in a variety of fields due to it has the ability to solve problems that cannot be handled by traditional technologies. The optimization of deep learning relevant to medical images is one of the most important recent advances in image analysis. Several developments have been done on Convolutional Neural Networks to achieve optimal accuracy and increase the learning speed. However, in this paper, we discuss the most recent innovations in convolutional neural networks within Classical method and Quantum method. We briefly provide a snapshot about the architecture, improvements, and principles of both (Classical and Quantum).
Renewable energy utilization in demand-side energy management system based on linear programming optimization algorithm
(American Institute of Physics, 2024) Almashhadani, Muna Kamel; Çevik, Mesut; Al-Jumaili, Saif; Alhanaf, Ahmed Sami; Al-Bhadely, Faraj Khlaf; Uçan, Osman Nuri
Demand-side management (DSM) is an effectual approach by coordinating utility management and routinely tracking energy usage, the intelligent grid assists in controlling energy demand and promotes its efficiency. However, the paper aims to utilize Linear programming optimization algorithms as an effective tool for managing energy demand and maximizing the use of renewable energy sources. These algorithms are able to estimate which is the best utilization of what resources are accessible and reduce consumption by describing the energy system as a collection of linear equations. The optimization system makes assumptions about the various energy costs when it will be high or low and modifies energy use accordingly. We applied different scenarios to assess the resiliency of the system. The simulation took into account a number of variables, including the weather, energy usage, and pricing fluctuations. MATLAB R2023a and Simulink provide an integrated platform with data analytics to build the proposed system and optimization model to minimize cost in MATLAB. Compared to other methods using various optimization algorithms as the binary orientation search algorithm (BOSA), cockroach swarm optimization (CSO), and the sparrow search algorithm (SSA) were applied to DSM methodology for a residential community with a primary focus on decreasing peak energy consumption results as in previous study was, BOSA has a lower standard deviation (0.8) compared to the other algorithms (1.7 for SSA and 1.3 for CSOA), making it more robust and superior, in addition to minimizing cost (5438.98 cents of USD (mean value) and 16.3% savings), the suggested approach is used for lowering electrical energy costs in a micro-grid system while maintaining their regular load and operating hours.
Transmission line loss determination of electricity by using convolutional neural network
(Institute of Electrical and Electronics Engineers Inc., 2022) Sadeq Al-Samkri, Elaf Hayder; Al-Jumaili, Saif; Noori, Harith Muthanna; Duru, Adil Deniz; Uçan, Osman Nuri
Businesses are concerned about energy losses. Electronic gadgets have become more prevalent as a result of their adoption. The frequency with which home electricity usage data is collected has grown, allowing for sophisticated data analysis that was previously unavailable. Indeed, adopting Smart Grid (SG) networks, which are freshly improved networks of linked devices, may considerably enhance the existing energy infrastructure's dependability, economy, and durability. The SG involves sharing a lot of data, including information on specific users' power use. And using this information, machine learning and deep learning algorithms may be able to detect power theft users reliably. This paper presented a Convolutional Neural Network (CNN)-based model for automated network-based vulnerability scanning that has excellent classification performance in many categories. Testing from iteration two to four iterations, this study examines research to discover the ideal configuration of the sequential model (SM) for categorization. The method is validated using a two-layer design, including an efficiency of 0.92, the whole first layer is composed of 128 nodes while the second level consists of 64 nodes. This allows for the development of a higher-level classifier for electrical signals, which may be employed in a number of applications. CNN was used to create electrical signal detectors, and SM was used to extract data from an electricity usage dataset. Furthermore, the number of features in the data set can be reduced using the Blue Monkey (BM) approach, and these results are then used to develop high-performance models. In this regard, the focus of this study has been on lowering the amount of needed features in the dataset in order to establish a rising classification algorithm for electrical signals. Experiments have applied the proposed systems' fantastic performance, with just 666 characteristics required to combine the CNN and BM methods. Comparative to 1035 traits when CNN was used alone. This shows that the CNN and BM models are better than the CNN model in terms of lowering sufficient know while maintaining the same reliability.
Using a novel method to improve various stages of machines in the power system
(Institute of Electrical and Electronics Engineers Inc., 2022) Ali, Mohammed Abdulkareem; Al-Jumaili, Saif; Noori, Harith Muthanna; Duru, Adil Deniz; Uçan, Osman Nuri
The aim of this research is to utilize the particle swarm approach to examine the coordinated design of the unified power flow controller (UPFC) and two power system stabilizers (PSS) in a multi-machine power system. The synchronized proposed control challenge has been presented as an optimal solution with an objective function and constraint equations to complete this goal. Period models in a multi-machine power system were used to assess the effectiveness of the proposed design under distinct operating situations. It has been shown that, even if both PSSs and UPFCs perform well in their own right, power networks stability is adversely affected by an unfavorable impact or negative interaction between controllers if they are not coordinated. Inter-Area oscillations have been efficiently dampened thanks to integrated layout.

Yazar "Al-Jumaili, Saif" seçeneğine göre listele

Sayfa Başına Sonuç

Sıralama seçenekleri