Yazar "Irmak, Emrah" seçeneğine göre listele
Listeleniyor 1 - 13 / 13
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe A Hybrid 2D Gaussian Filter and Deep Learning Approach with Visualization of Class Activation for Automatic Lung and Colon Cancer Diagnosis(Sage Publications Inc, 2024) Turk, Omer; Acar, Emrullah; Irmak, Emrah; Yilmaz, Musa; Bakis, EnesCancer is a significant public health issue due to its high prevalence and lethality, particularly lung and colon cancers, which account for over a quarter of all cancer cases. This study aims to enhance the detection rate of lung and colon cancer by designing an automated diagnosis system. The system focuses on early detection through image pre-processing with a 2D Gaussian filter, while maintaining simplicity to minimize computational requirements and runtime. The study employs three Convolutional Neural Network (CNN) models-MobileNet, VGG16, and ResNet50-to diagnose five types of cancer: Colon Adenocarcinoma, Benign Colonic Tissue, Lung Adenocarcinoma, Benign Lung Tissue, and Lung Squamous Cell Carcinoma. A large dataset comprising 25 000 histopathological images is utilized. Additionally, the research addresses the need for safety levels in the model by using Class Activation Mapping (CAM) for explanatory purposes. Experimental results indicate that the proposed system achieves a high diagnostic accuracy of 99.38% for lung and colon cancers. This high performance underscores the effectiveness of the automated system in detecting these types of cancer. The findings from this study support the potential for early diagnosis of lung and colon cancers, which can facilitate timely therapeutic interventions and improve patient outcomes.Öğe A novel deep convolutional neural network model for COVID-19 disease detection(Institute of Electrical and Electronics Engineers Inc., 2020) Irmak, EmrahThe novel coronavirus, generally known as COVID19, is a new type of coronavirus which first appeared in Wuhan Province of China in December 2019. The biggest impact of this new coronavirus is its very high contagious feature which brings the life to a halt. As soon as data about the nature of this dangerous virus are collected, the research on the diagnosis of COVID-19 has started to gain a lot of momentum. Today, the gold standard for COVID-19 disease diagnosis is typically based on swabs from the nose and throat, which is time-consuming and prone to manual errors. The sensitivity of these tests are not high enough for early detection. These disadvantages show how essential it is to perform a fully automated framework for COVID-19 disease diagnosis based on deep learning methods using widely available X-ray protocols. In this paper, a novel, powerful and robust Convolutional Neural Network (CNN) model is designed and proposed for the detection of COVID-19 disease using publicly available datasets. This model is used to decide whether a given chest X-ray image of a patient has COVID-19 or not with an accuracy of 99.20%. Experimental results on clinical datasets show the effectiveness of the proposed model. It is believed that study proposed in this research paper can be used in practice to help the physicians for diagnosing the COVID-19 disease. © 2020 IEEE.Öğe A novel implementation of deep-learning approach on malaria parasite detection from thin blood cell images(2021) Irmak, EmrahMalaria is known as an acute febrile disease caused by the bite of female Anopheles mosquitoes, and it manifests itself with symptoms such as headache, fever, chills, vomiting, and fatigue. The diagnosis of malaria is still based on manual identification of Plasmodium parasitized cells in microscopic examinations of blood cells known as parasite based microscopy diagnostic testing. The accuracy of this manual diagnosis meth , is clearly affected by the level of microscopists experience, which makes this diagnosis method susceptible to manual error and time consuming. Diagnoses of diseases made using deep learning methods have had great repercussions in the medical world, especially in recent years; and this indicate, that the diagnosis of malaria can also be achieved by deep learning methods. On the basis of this fact, this paper presents a novel deep-learning-based malaria disease detection technique. A convolutional neural network (CNN) architecture, which has 20 weighted layers is designed and proposed to identify parasitized microscopic images from uninfected microscopic images. A total of 27,558 thin blood cell images were used to train and test the CNN model, and 95.28% overall accuracy was obtained. The experimental results on large clinical dataset show the effectiveness of the piposed deep learning method for malaria disease detection.Öğe Consistency and Comparison of Monomodal Multi-Temporal Medical Image Registration-Segmentation and Mathematical Model for Glioblastoma Volume Progression(2020) Irmak, EmrahTumor volume progression analysis and tumor volume measurement are very common tasks in cancer research and image processing fields. Tumor volume measurement can be carried out in two ways. The first way is to use different mathematical formulas and the second way is to use image registration method. In this paper, using 3D medical image registration-segmentation algorithm, multiple scans of MR images of a patient who has brain tumor are registered with different MR images of the same patient acquired at a different time so that growth of the tumor inside the patient's brain can be investigated. Tumor volume progression analysis and tumor volume measurement are performed using image registration technique and the results are compared with the results of tumor volume measurement by mathematical formulas. For the first patient, grown brain tumor volume is found to be 10345 mm³, diminished brain tumor volume is found to be 15278 mm³ and unchanged brain tumor volume is found to be 20876 mm³. Numerical results obtained by image registration model proves that medical image - registration method is not only between the true ranges but also is very close to the best mathematical formula. Medical image registration-segmentation are implemented to 19 patients and satisfactory results are obtained The results are compared with the results obtained from mathematical methods. An advantageous point of medical image registration-segmentation method over mathematical models for brain tumor investigation is that grown, diminished, and unchanged brain tumor parts of the patients are investigated and computed on an individual basis in a three - dimensional (3D) manner within the time.Öğe COVID-19 disease diagnosis from paper-based ECG trace image data using a novel convolutional neural network model(2022) Irmak, EmrahClinical reports show that COVID-19 disease has impacts on the cardiovascular system in addition to the respiratory system. Available COVID-19 diagnostic methods have been shown to have limitations. In addition to current diagnostic methods such as low-sensitivity standard RT-PCR tests and expensive medical imaging devices, the development of alternative methods for the diagnosis of COVID-19 disease would be benefcial for control of the COVID-19 pandemic. Further, it is important to quickly and accurately detect abnormalities caused by COVID-19 on the cardiovascular system via ECG. In this study, the diagnosis of COVID-19 disease is proposed using a novel deep Convolutional Neural Network model by using only ECG trace images created from ECG signals of COVID-19 infected patients based on the abnormalities caused by the COVID-19 virus on the cardiovascular system. An overall classifcation accuracy of 98.57%, 93.20%, 96.74% and AUC value of 0.9966, 0.9771, 0.9905 is achieved for COVID-19 vs. Normal, COVID-19 vs. Abnormal Heartbeats, COVID-19 vs. Myocardial Infarction binary classifcation tasks, respectively. In addition, an overall classifcation accuracy of 86.55% and 83.05% is achieved for COVID-19 vs. Abnormal Heartbeats vs. Myocardial Infarction and Normal vs. COVID-19 vs. Abnormal Heartbeats vs. Myocardial Infarction multi-classifcation tasks. This study is believed to have great potential to speed up the diagnosis and treatment of COVID-19 patients, saving clinicians time and facilitating the control of the pandemic.Öğe COVID-19 disease severity assessment using CNN model(2021) Irmak, EmrahDue to the highly infectious nature of the novel coronavirus (COVID-19) disease, excessive number of patients waits in the line for chest X-ray examination, which overloads the clinicians and radiologists and negatively affects the patient's treatment, prognosis and control of the pandemic. Now that the clinical facilities such as the intensive care units and the mechanical ventilators are very limited in the face of this highly contagious disease, it becomes quite important to classify the patients according to their severity levels. This paper presents a novel implementation of convolutional neural network (CNN) approach for COVID-19 disease severity classification (assessment). An automated CNN model is designed and proposed to divide COVID-19 patients into four severity classes as mild, moderate, severe, and critical with an average accuracy of 95.52% using chest X-ray images as input. Experimental results on a sufficiently large number of chest X-ray images demonstrate the effectiveness of CNN model produced with the proposed framework. To the best of the author's knowledge, this is the first COVID-19 disease severity assessment study with four stages (mild vs. moderate vs. severe vs. critical) using a sufficiently large number of X-ray images dataset and CNN whose almost all hyper-parameters are automatically tuned by the grid search optimiser.Öğe ESO-based Backstepping Control of DC-DC Buck Converter Under Mismatched load Disturbance(Alanya Alaaddin Keykubat University, 2023) Uslu, Ümit Akın; Irmak, EmrahThe DC–DC power converter play a critical role to provide stable DC output voltage, however which is subject to various uncertainties and disturbance in supplying the sensitive loads. This paper propose a composite backstepping control strategy with extended state observer (ESO) for buck converter. Firstly, a backstepping control function is constructed to derive an inner current loop reference assuming load disturbance is known, which renders global stability of the system. An ESO is designed to estimate the mismatched load current and feedforward it to the backstepping controller to obtain disturbance rejection. Quantitative selection of control and observer gains are provided under highly nonlinear relationship of system dynamics. Rigorous stability of proposed scheme are poved with analysis of robustness. Finally, simulation results illustrate the effectiveness of the proposed control scheme in the presence of load disturbance and uncertaintiesÖğe Gerilim Kaynaklı Dönüştürücülerin Gürbüz Doğrudan Güç Akış Kontrolü(Alanya Alaaddin Keykubat Üniversitesi, 2024) Uslu, Ümit Akın; Irmak, EmrahIn this study, a disturbance observer-based power control system is developed for voltage source converters (VSC) to achieve smooth power delivery to the grid. Firstly, modeling of grid connected converter which is used for power delivery in terms of frequency and current dynamics is executed under consideration of modelling errors and uncertainties. This disturbance effects are mainly consist of frequency and amplitude variations, output impedance aging, large dc-link voltage ripple A second order nonlinear observer is integrated into the designed control to reject with model uncertainty and disturbances. Due to the objective of the proposed controller is power regulation, there is no need a voltage and current compensator. Comperative simulations are carried out to verify the robustness of the proposed controller. Also effectiveness of the proposed approach is tested under different grid scenarios (e.g., weak grid, dc-link variation, frequency deviation)Öğe Implementation of convolutional neural network approach for COVID-19 disease detection(American Physiological Society, 2020) Irmak, EmrahIn this paper, two novel, powerful, and robust convolutional neural network (CNN) architectures are designed and proposed for two different classification tasks using publicly available data sets. The first architecture is able to decide whether a given chest X-ray image of a patient contains COVID-19 or not with 98.92% average accuracy. The second CNN architecture is able to divide a given chest X-ray image of a patient into three classes (COVID-19 versus normal versus pneumonia) with 98.27% average accuracy. The hyperparameters of both CNN models are automatically determined using Grid Search. Experimental results on large clinical data sets show the effectiveness of the proposed architectures and demonstrate that the proposed algorithms can overcome the disadvantages mentioned above. Moreover, the proposed CNN models are fully automatic in terms of not requiring the extraction of diseased tissue, which is a great improvement of available automatic methods in the literature. To the best of the author’s knowledge, this study is the first study to detect COVID-19 disease from given chest X-ray images, using CNN, whose hyperparameters are automatically determined by the Grid Search. Another important contribution of this study is that it is the first CNN-based COVID-19 chest X-ray image classification study that uses the largest possible clinical data set. A total of 1,524 COVID-19, 1,527 pneumonia, and 1524 normal Xray images are collected. It is aimed to collect the largest number of COVID-19 X-ray images that exist in the literature until the writing of this research paper. © 2020 the American Physiological Society.Öğe Investigation of Tribocorrosion Properties of Titanium Implant used in Orthopedics(Ieee, 2022) Irmak, Emrah; Ugurlu, Bilal; Incesu, AlperA titanium implant is a mechanical system that undergoes tribocorrosion at the interface between the implant and the abutment alloy, where material degradation is common. When various titanium implant materials, which are used to fix the bones broken by trauma or to replace a joint or bones, such as osteoporosis, are exposed to a simulated body fluid (SBF), the effects of wear and corrosion phenomena in these titanium implant materials are observed to be determined quantitatively. For the samples extracted from Titanium-Ti6Al4V screws belonging to two different companies (A and B), very important parameters for tribocorrosion such as wear rate, depth of wear and corrosion rate were determined experimentally in the laboratory environment. The results were examined how suitable titanium is to resist material loss in body implants due to both wear and corrosion. In addition, it was determined that the wear rates obtained from the sample belonging to company B were more stable than those of company A, and the corrosion rates (0.0047 mm/year) obtained from the sample of company A were lower than those of company B.Öğe Measurement of important tribocorrosion properties of titanium implant and assessment with digital image processing(2025) Irmak, Emrah; Körpe, EnisTitanium implants are mechanical systems where tribocorrosion is frequently observed at the interface between the implant and the abutment alloy. In this paper, important parameters in terms of tribocorrosion such as abrasion coefficient, abrasion volume loss and corrosion rate were determined experimentally in a laboratory environment by preparing a sufficient number of samples obtained from the field for this material (Titanium-Ti6Al4V), which is actively used in surgeries. Comprehensive analysis of these mechanical systems in body-like environments contributes to a better understanding of the material loss caused by abrasion and corrosion interactions occurring at the interface between the implant and the abutment alloy. The samples were subjected to dry sliding wear with the pin-on-disc system in accordance with the relevant standards for a certain number of cycles, during which abrasion volume loss and friction coefficient were measured simultaneously. The results of these experiments were examined to evaluate the degree to which titanium is resistant to material loss due to abrasion and corrosion in body implants. It was found that the abrasion coefficient decreased by 51% when 10 N load was applied by 22% when 20 N load was applied, and by 2% when 30 N load was applied. The samples screws were exposed to 15% more corrosive abrasion and it was found that they had 6% higher corrosion rate in electrochemical corrosion test. Additionally, the morphological features of the abraided and corroded surfaces of Ti6Al4V alloy were analyzed and interpreted using scanning electron microscopy (SEM) and digital image processing techniques.Öğe Measurement of important tribocorrosion properties of titanium implant and assessment with digital image processing(Pamukkale Univ, 2025) Irmak, Emrah; Korpe, EnisTitanium implants are mechanical systems where tribocorrosion is frequently observed at the interface between the implant and the abutment alloy. In this paper, important parameters in terms of tribocorrosion such as abrasion coefficient, abrasion volume loss and corrosion rate were determined experimentally in a laboratory environment by preparing a sufficient number ofsamples obtained from the field for this material (Titanium-Ti6Al4V), which is actively used in surgeries. Comprehensive analysis of these mechanical systems in body-like environments contributes to a better understanding of the material loss caused by abrasion and corrosion interactions occurring at the interface between the implant and the abutment alloy. The samples were subjected to dry sliding wear with the pin-on-disc system in accordance with the relevant standards for a certain number of cycles, during which abrasion volume loss and friction coefficient were measured simultaneously. The results of these experiments were examined to evaluate the degree to which titanium is resistant to material loss due to abrasion and corrosion in body implants. It was found that the abrasion coefficient decreased by 51% when 10 N load was applied by 22% when 20 N load was applied, and by 2% when 30 N load was applied. The samples screws were exposed to 15% more corrosive abrasion and it was found that they had 6% higher corrosion rate in electrochemical corrosion test. Additionally, the morphological features of the abraided and corroded surfaces of Ti6Al4V alloy were analyzed and interpreted using scanning electron microscopy (SEM) and digital image processing techniques.Öğe Multi-classification of brain tumor MRI images using deep convolutional neural network with fully optimized framework(2021) Irmak, EmrahBrain tumor diagnosis and classification still rely on histopathological analysis of biopsy specimens today. The current method is invasive, time-consuming and prone to manual errors. These disadvantages show how essential it is to perform a fully automated method for multi-classification of brain tumors based on deep learning. This paper aims to make multi-classification of brain tumors for the early diagnosis purposes using convolutional neural network (CNN). Three different CNN models are proposed for three different classification tasks. Brain tumor detection is achieved with 99.33% accuracy using the first CNN model. The second CNN model can classify the brain tumor into five brain tumor types as normal, glioma, meningioma, pituitary and metastatic with an accuracy of 92.66%. The third CNN model can classify the brain tumors into three grades as Grade II, Grade III and Grade IV with an accuracy of 98.14%. All the important hyper-parameters of CNN models are automatically designated using the grid search optimization algorithm. To the best of author's knowledge, this is the first study for multi-classification of brain tumor MRI images using CNN whose almost all hyper-parameters are tuned by the grid search optimizer. The proposed CNN models are compared with other popular state-of-the-art CNN models such as AlexNet, Inceptionv3, ResNet-50, VGG-16 and GoogleNet. Satisfactory classification results are obtained using large and publicly available clinical datasets. The proposed CNN models can be employed to assist physicians and radiologists in validating their initial screening for brain tumor multi-classification purposes.
