Sürdürülebilir Kalkınma Amaçları
Yazarlar
- Melek, Mesut(7)
- Bozkurt, Ferhat(2)
- Celik, Kemal(2)
- Gökrem, Levent(2)
- Günen, Mehmet Akif(2)
- Kaya, Serhat(2)
- Kayikcioglu, Temel(2)
- Khoshelham, Kourosh(2)
- Manshouri, Negin(2)
- Melek, Negin(2)
- Reis, Hatice Catal(2)
- Ture, Hayati(2)
- Turk, Veysel(2)
- Almasani, Heba(1)
- Aydin, Sefa(1)
- Aydın, Sefa(1)
- Balcik, Filiz Bektas(1)
- Besdok, Erkan(1)
- Büşra Geliç(1)
- CELIK, Ayse(1)
- CELIK, Fatih(1)
- CELIK, Kemal(1)
- Celik, Fatih(1)
- Dogan, Ramazan Ozgur(1)
- Goksel, Cigdem(1)
- Incir, Ramazan(1)
- Kaylkcloglu, Temel(1)
- Kobya, Mehmet Emin(1)
- Kul, Sibel(1)
- Maleki, Masoud(1)
- Teymur, Aykut(1)
- Topaloglu, Raziye Hale(1)
- İncir, Ramazan(1)
Yayınlar
- MakaleYalnız künye2026Scopus SJRQ2A Dense and Sparse Attention in Vision Transformer for Remote Sensing Image Classification
Celik, Kemal
Satellite imagery plays a crucial role in various applications, such as environmental monitoring, urban planning, and disaster response. Previously, Convolutional neural networks have played a crucial role in improving remote sensing scene classification by effectively capturing spatial features. However, they often struggle with long-range dependencies and complex spectral patterns in satellite images. Recently, transformer-based architectures, initially designed for natural language processing, have been adopted in computer vision to leverage self-attention mechanisms for capturing long-range dependencies. To effectively extract the long-range dependencies and complex patterns, this paper proposes a novel method combining dense and sparse attention mechanisms to enhance remote sensing scene classification in different scales. The dense attention mechanisms capture comprehensive connections within satellite images, followed by sparse attention mechanisms applied at both tile and patch scale levels. This multi-scale attention strategy allows the model to efficiently focus on important spatial and spectral features while managing computational complexity. Extensive experimental evaluations on three benchmark datasets, such as Merced, AID, and Optimal-31, demonstrate significant improvements in classification accuracy, achieving 98.9, 97.5, and 97.8, respectively. Experimental results show that the proposed framework outperforms CNN and Transformer baselines, enhancing the accuracy and robustness of remote sensing scene classification for geospatial analysis.
AtıfScopus2WoS2OpenAlex1
İncele - MakaleYalnız künye2026Scopus SJRQ2Multi level throttled attention for accurate and efficient classification of retinal diseases in OCT images
Celik, Kemal; Celik, Fatih
The classification of retinal diseases using Optical Coherence Tomography images has gained significant attention in recent years. However, due to the heterogeneity of retinal disorders, achieving high classification accuracy remains challenging. While deep learning has improved the identification of various ophthalmic conditions, accurate and early diagnosis is still critical for effective clinical decision-making. Additionally, reducing computational complexity is essential for building scalable diagnostic systems. To address these challenges, we propose a novel Multi-level Throttled Attention (MLTA) model that first applies attention to extract important features, followed by throttling to filter out redundant or less useful features. This process is applied at multiple stages, operating on both Convolutional Neural Network layers and Transformer architectures, and then fuses them finally. This approach optimizes feature selection by preserving the most informative retinal features while discarding redundant ones from multi-level feature maps. As a result, the model enhances classification performance while reducing computational overhead. We evaluate our proposed model on two benchmark OCT datasets: OCT2017 and OCTID, achieving accuracy scores of 99.59 ± 0.04% and 99.71 ± 0.03%, respectively. Our method outperforms existing CNN-based and Vision Transformer-based models, demonstrating superior accuracy and efficiency through this hierarchical attention.
- MakaleYalnız künye2025Scopus SJRQ3Golden Distance: A New and Comprehensive Metric Definition Study Facilitating Classification Performance Evaluations
Melek, Mesut; Melek, Negin
Machine/deep learning has great potential in many classification studies that make our lives easier. Performance evaluation of the classification algorithm is of great importance for making accurate predictions in machine learning, which, itself, learns from data without any human intervention. In these learning models, high-dimensional data in classes with complex features belonging to balanced or unbalanced distributions are analysed with different classification algorithms and take their place in the decision stage. While the difficulties in implementing classification algorithms continue today, evaluating the superiority of developed systems over each other continues to be both a time-consuming and very difficult process. In models proposed in the literature, only a single parameter, such as accuracy, is not sufficient to determine model adequacy. In this context, among different classification algorithm performance parameters, some metrics such as sensitivity, specificity, F-measure (F1), Jaccard index, area under the curve, and kappa coefficient are used for system evaluation. As these parameters are placed in the analysis tables, the study should be evaluated at length by including large tables in the article. Instead of using various parameters that make performance analysis difficult in system evaluation, this study defines a new Golden Distance-based parameter that summarizes the model comprehensively from all dimensions. To understand this new metric’s working principle, a detailed comparative analysis of research conducted on three datasets was performed. Based on the final evaluation, the proposed inclusive metric demonstrated successful and fair behaviour in all conditions. Thus, a fair and simple comparison will be achieved with a single metric, and long analyses can be avoided by presenting many performance parameters in systems’ superiority indicator tables.
AtıfScopus9WoS10OpenAlex12İncele - MakaleAçık Erişim2025Scopus SJRQ2Intersession Robust Hybrid Brain–Computer Interface: Safe and User-Friendly Approach with LED Activation Mechanism
Aydın, Sefa; Melek, Mesut; Gökrem, Levent
This study introduces a hybrid Brain–Computer (BCI) system with a robust and secure activation mechanism between sessions, aiming to minimize the negative effects of visual stimulus-based BCI systems on user eye health. The system is based on the integration of Electroencephalography (EEG) signals and Electrooculography (EOG) artefacts, and includes an LED stimulus operating at a frequency of 7 Hz for safe activation and objects moving in different directions. While the LED functions as an activation switch that reduces visual fatigue caused by traditional visual stimuli, moving objects provide command generation depending on the user’s intention. In order to evaluate the stability of the system against physiological and psychological conditions, data were collected from 15 participants in two different sessions. The Correlation Alignment (CORAL) method was applied to the data to reduce the variance between sessions and to increase stability. A Bootstrap Aggregating algorithm was used in the classification processes, and with the CORAL method, the system accuracy rate was increased from 81.54% to 94.29%. Compared to similar BCI approaches, the proposed system offers a safe activation mechanism that effectively adapts to users’ changing cognitive states throughout the day by reducing visual fatigue, despite using a low number of EEG channels, and demonstrates its practicality and effectiveness by performing on par or superior to other systems in terms of high accuracy and robust stability.
AtıfScopus2WoS1OpenAlex2
İncele - BildiriYalnız künye2025Improving Skin Lesion Classification with Pretrained Deep Learning Models
Almasani, Heba; Ture, Hayati
Skin cancer is the most common type of cancer. Cancer can be defined as a disease characterized by the uncontrolled division and spread of cells. As with other types of cancer, early diagnosis and treatment play a crucial role in combating skin cancer. Since diagnosis is made through visual examination, factors such as the level of expertise and image artifacts negatively affect diagnostic accuracy and slow down the process.Computer-aided applications used to analyze such images provide faster and more accurate results. Additionally, these methods eliminate factors that reduce diagnostic accuracy during the classification and segmentation of skin lesions.This study classifies skin cancer to benign or malignant using deep learning and machine learning methods. In particular, Convolutional Neural Networks (CNNs) have shown promising results in medical image classification. The study evaluates six pre-trained CNN models - VGG16, EfficientNetB7, EfficientNetV2L, DenseNet, ConvNeXtBase and MobileNet - for binary skin lesion classification using the ISIC 2019 dataset. The study first preprocesses and balances the data to enhance performance. The results indicate that the ConvNeXtBase model achieved the highest success with accuracy rate of 84.63% after balancing the data that we have.
AtıfScopus5WoS1OpenAlex5İncele - MakaleAçık Erişim2025Scopus SJRQ2A Safe and Efficient Brain–Computer Interface Using Moving Object Trajectories and LED-Controlled Activation
Aydin, Sefa; Melek, Mesut; Gökrem, Levent
Nowadays, brain–computer interface (BCI) systems are frequently used to connect individuals who have lost their mobility with the outside world. These BCI systems enable individuals to control external devices using brain signals. However, these systems have certain disadvantages for users. This paper proposes a novel approach to minimize the disadvantages of visual stimuli on the eye health of system users in BCI systems employing visual evoked potential (VEP) and P300 methods. The approach employs moving objects with different trajectories instead of visual stimuli. It uses a light-emitting diode (LED) with a frequency of 7 Hz as a condition for the BCI system to be active. The LED is assigned to the system to prevent it from being triggered by any involuntary or independent eye movements of the user. Thus, the system user will be able to use a safe BCI system with a single visual stimulus that blinks on the side without needing to focus on any visual stimulus through moving balls. Data were recorded in two phases: when the LED was on and when the LED was off. The recorded data were processed using a Butterworth filter and the power spectral density (PSD) method. In the first classification phase, which was performed for the system to detect the LED in the background, the highest accuracy rate of 99.57% was achieved with the random forest (RF) classification algorithm. In the second classification phase, which involves classifying moving objects within the proposed approach, the highest accuracy rate of 97.89% and an information transfer rate (ITR) value of 36.75 (bits/min) were achieved using the RF classifier.
AtıfScopus4WoS5OpenAlex3İncele - BildiriAçık Erişim2024A Study on the Segmentation and Classification of Diabetic Retinopathy Images Using the K-Means Clustering Method
Incir, Ramazan; Bozkurt, Ferhat
Diabetic Retinopathy (DR) is a retinal disease caused by diabetes, representing one of the most prevalent causes of vision loss affecting millions of people worldwide. Swift detection and treatment of this condition are crucial for preventing the disease. Various deep learning and machine learning algorithms have been employed for disease detection and classification, often overlooking the data preprocessing stage. In the data preprocessing phase of this study, segmentation of important lesions, such as hard exudates, was conducted using the K-Means clustering method. The identified lesions were highlighted on the original images. The resulting dataset was then classified using pre-trained architectures, namely EfficientNetV2-M, ResNet50, MobileNet, and DenseNet121. After training on the APTOS dataset, the EfficientNetV2-M model achieved an accuracy of 95.16%. The classification results indicated the contribution of the lesion highlighting process during data preprocessing to the overall classification accuracy.
AtıfWoS1OpenAlex3İncele - MakaleAçık Erişim2024Scopus SJRQ1A study on effective data preprocessing and augmentation method in diabetic retinopathy classification using pre-trained deep learning approaches
İncir, Ramazan; Bozkurt, Ferhat
High glucose levels in the blood not only damage different tissues and organs of the body, but also cause adverse effects on the eye. This condition is called diabetic retinopathy (DR). DR can cause blurred vision, darkening of the field of vision, and severe vision loss. The number of people infected with the disease is increasing in our country and worldwide. The time-consuming physician check-ups and the presence of small lesions indicate the need to develop diagnostic systems. Deep learning-based applications have become the trend for diagnosing and grading diseases from images. This study aims to create a meaningful and sufficient dataset using effective data preprocessing and affine transformation techniques in diabetic retinopathy classification. In this study, classification was performed using seven different pre-trained deep learning architectures. An experimental study of each technique was performed on the EyePACS dataset. An overfitting problem was encountered in the experimental results with the original data set. Thus, data preprocessing and data augmentation processes were carried out in order to eliminate overfitting by considering the imbalance between classes in the dataset. The classification performance obtained from each architecture was observed according to performance metrics of precision, recall, F1 Score, accuracy, and loss. In this study, the best performance was achieved with 97.65% test accuracy with the proposed EfficientNetV2-M network model.
AtıfScopus32WoS21OpenAlex34İncele - MakaleAçık Erişim2024Scopus SJRQ1Enhancing brain tumor classification through ensemble attention mechanism
CELIK, Fatih; CELIK, Kemal; CELIK, Ayse
Brain tumors pose a serious threat to public health, impacting thousands of individuals directly or indirectly worldwide. Timely and accurate detection of these tumors is crucial for effective treatment and enhancing the quality of patients’ lives. The widely used brain imaging technique is magnetic resonance imaging, the precise identification of brain tumors in MRI images is challenging due to the diverse anatomical structures. This paper introduces an innovative approach known as the ensemble attention mechanism to address this challenge. Initially, the approach uses two networks to extract intermediate- and final-level feature maps from MobileNetV3 and EfficientNetB7. This assists in gathering the relevant feature maps from the different models at different levels. Then, the technique incorporates a co-attention mechanism into the intermediate and final feature map levels on both networks and ensembles them. This directs attention to certain regions to extract global-level features at different levels. Ensemble of attentive feature maps enabling the precise detection of various feature patterns within brain tumor images at both model, local, and global levels. This leads to an improvement in the classification process. The proposed system was evaluated on the Figshare dataset and achieved an accuracy of 98.94%, and 98.48% for the BraTS 2019 dataset which is superior to other methods. Thus, it is robust and suitable for brain tumor detection in healthcare systems.
AtıfScopus42WoS21OpenAlex42İncele - Yüksek Lisans TeziAçık Erişim2023Eczane envanterinin EDAS yöntemiyle sınıflandırılması
Büşra Geliç
Sağlık kurumları içinde hizmet sunan sektörlerden önemli birisi de eczanelerdir. Eczanelerin daha iyi hizmet sunabilmesi için stok kontrolünün doğru ve düzenli yapılması gerekmektedir. Eczane sayılarındaki artış, ilaç çeşitliliği, ilaç fiyatlarındaki değişim gibi nedenler envanter yönetimini zorunlu kılmaktadır. Envanter yönetimi, talep edilen malzemeye doğru miktarda ve zamanda ulaşma sürecidir. Malzemeleri sınıflandırmak bu süreci kolaylaştırabilir. Envanter sınıflandırmada çoğunlukla ABC yöntemi kullanılmaktadır. Bu yöntem tek bir kriteri dikkate alırken günümüzde ilaçların sınıflandırmasında çok sayıda faktörün önemi olduğu için çok kriterli karar verme teknikleri envanter sınıflandırma probleminde yaygın olarak kullanılmaktadır. Bu çalışmada eczane envanter sınıflandırma problemi gerçek verilerle ele alınmıştır. Eczane envanterinde yer alan 1139 adet ilaç maliyet, stok ve kar yüzdeleri kriterleri dikkate alınarak EDAS tekniği ile elde edilen ağırlıklar eşit kabul edilerek sınıflandırma yapılmıştır. Sınıflandırma sonucunda A grubunda 114, C grubunda 683 ve B grubunda 342 ilaç yer almıştır. EDAS yöntemiyle elde edilen ağırlık setlerinin doğruluğunu teyit etmek için duyarlılık analizinde sonuçları nasıl etkilediğine bakılmıştır. Analiz sonucunda envanter sınıflandırma problemlerinde istatistik tabanlı ağırlıklandırmalarla sınıflandırma yapılsa da değişiklik olduğu tespit edildiği için uzman görüşü alınması gerektiği sonucuna varılmıştır. Anahtar Kelimeler: BWM yöntemi, Eczane envanter sınıflandırma, EDAS yöntemi, Stok yönetimi
- BildiriYalnız künye2023Shape of Masses in Contrast Mammography Images and First Order Statistical Features Classification Based on
Kobya, Mehmet Emin; Ture, Hayati; Dogan, Ramazan Ozgur; Kayikcioglu, Temel; Teymur, Aykut; Kul, Sibel
Mammography is a valuable tool in the diagnosis and staging of primary breast cancer. Contrast-enhanced mammography (CEM) combines an iodinated contrast agent with traditional mammography to increase diagnostic accuracy, especially in women with more dense parenchymal background patterns. In this study, a Decision Support System (DSS) is proposed that classifies masses using radiomic features obtained from masks of benign and malignant masses in images obtained with the CEM technique. An open-access database containing 2006 CEM images was used in this study. Radiomic features obtained from these images were used to identify differences between benign and malignant masses. The DSS uses these features to classify masses using a decision tree algorithm. This study highlights the importance of the use of iodinated contrast agents to improve diagnostic accuracy in women with more dense parenchymal background patterns. Additionally, it emphasizes the potential benefits of a DSS, in combination with radiomic features, in the diagnosis and staging of breast cancer. In conclusion, this study demonstrates that the use of radiomic features in mammography images obtained with the CEM technique is an effective way to classify benign and malignant masses.The use of this technology could be an important tool in the early diagnosis and accurate classification of breast cancer.
- MakaleAçık Erişim2023Effect of denoising methods for hyperspectral images classification: DnCNN, NGM, CSF, BM3D and Wiener
Günen, Mehmet Akif; Besdok, Erkan
Hyperspectral images are widely used for land use/cover analysis in remote sensing due to their rich spectral information. However, these data often suffer from noise caused by various factors such as random and systematic errors, making them less useful for end-users. In this study, denoising methods (i.e., DnCNN, NGM, CSF, BM3D, and Wiener) for hyperspectral images were compared using the Pavia University hyperspectral dataset with four different noise types: Gaussian, Salt & Pepper, Poisson, and Speckle. After denoising, the k-nearest neighbor method was used to classify the image, and statistical and visual performance comparisons were performed on the classified data. Six performance metrics-Accuracy, Sensitivity, Specificity, Precision, F-Score, and G-Mean-were employed to compare the outcomes qualitatively. The findings demonstrate that DnCNN and BM3D have the best outcome performance for all four noise types. Due to their lack of sensitivity and specificity, the CSF and Wiener approaches had low performance for particular noise sources. For all noise types, the NGM approach had the worst results. The validated instruments not provide effective results when it came to denoising Salt & Pepper noise, but they managed to produce outstanding results when it came to denoising Poisson noise. In order to enhance the quality and usability of hyperspectral images for land use/cover analysis, this study emphasizes the significance of choosing an effective denoising technique.
AtıfScopus5OpenAlex5İncele - MakaleYayıncıda açık erişim2023Scopus SJRQ1MediNet: transfer learning approach with MediNet medical visual database
Reis, Hatice Catal; Turk, Veysel; Khoshelham, Kourosh; Kaya, Serhat
The rapid development of machine learning has increased interest in the use of deep learning methods in medical research. Deep learning in the medical field is used in disease detection and classification problems in the clinical decision-making process. Large amounts of labeled datasets are often required to train deep neural networks; however, in the medical field, the lack of a sufficient number of images in datasets and the difficulties encountered during data collection are among the main problems. In this study, we propose MediNet, a new 10-class visual dataset consisting of Rontgen (X-ray), Computed Tomography (CT), Magnetic Resonance Imaging (MRI), Ultrasound, and Histopathological images such as calcaneal normal, calcaneal tumor, colon benign colon adenocarcinoma, brain normal, brain tumor, breast benign, breast malignant, chest normal, chest pneumonia. AlexNet, VGG19-BN, Inception V3, DenseNet 121, ResNet 101, EfficientNet B0, Nested-LSTM + CNN, and proposed RdiNet deep learning algorithms are used in the transfer learning for pre-training and classification application. Transfer learning aims to apply previously learned knowledge in a new task. Seven algorithms were trained with the MediNet dataset, and the models obtained from these algorithms, namely feature vectors, were recorded. Pre-training models were used for classification studies on chest X-ray images, diabetic retinopathy, and Covid-19 datasets with the transfer learning technique. In performance measurement, an accuracy of 94.84% was obtained in the traditional classification study for the InceptionV3 model in the classification study performed on the Chest X-Ray Images dataset, and the accuracy was increased 98.71% after the transfer learning technique was applied. In the Covid-19 dataset, the classification success of the DenseNet121 model before pre-trained was 88%, while the performance after the transfer application with MediNet was 92%. In the Diabetic retinopathy dataset, the classification success of the Nested-LSTM + CNN model before pre-trained was 79.35%, while the classification success was 81.52% after the transfer application with MediNet. The comparison of results obtained from experimental studies observed that the proposed method produced more successful results. Graphical abstract: [Figure not available: see fulltext.]
AtıfScopus16WoS14OpenAlex18İncele - MakaleYalnız künye2022Scopus SJRQ1Performance comparison of deep learning and machine learning methods in determining wetland water areas using EuroSAT dataset
Günen, Mehmet Akif
Wetlands are critical to the ecology because they maintain biodiversity and provide home for a variety of species. Researching, mapping, and conservation of wetlands is a challenging and time-consuming process. Because they produce temporal and geographical information, remote sensing and photogrammetric approaches are useful tools for analyzing and managing wetlands. In this study, the water areas of five different wetlands obtained with Sentinel-2 images in Turkey were classified. Although obtaining large amounts of high-dimensional dataset labeled for various land types is costly, it is a significant advantage to use it after model training in a wide range of applications. In this paper, the EuroSAT dataset was used in the validation process. Proposed deep learning–based 1D convolutional neural networks (CNN) and traditional machine learning methods (i.e., support vector machine, linear discriminant analysis, K-nearest neighborhood, canonical correlation forests, and AdaBoost.M1) were compared quantitatively (i.e., accuracy, recall, precision, specificity, F-score, and image quality assessment metrics) and qualitatively. Finally, pairwise comparison was made with chi-square-based McNemar’s test. There is a statistical difference between 1D CNN and machine learning method (except the support vector machine vs linear discriminant analysis in Test 1 area). CNN models outperform machine learning algorithms in terms of non-linear function approximation and the ability to extract and articulate data features. Since 1D CNNs can process data in a highly complex and unique feature space, they are very successful in segmenting strongly related and highly correlated discrete signals. It also has advantages over machine learning methods for water body extraction in that it can be integrated with sophisticated image pre-processing and standardization tools, is less susceptible to low-level random noise, and provides shift in variations and contrast-invariant image local transforms.
AtıfScopus74WoS53OpenAlex72İncele - MakaleYalnız künye2022Scopus SJRQ2Detection of 2D and 3D Video Transitions Based on EEG Power
Manshouri, Negin; Melek, Mesut; Kaylkcloglu, Temel
Despite the long and extensive history of 3D technology, it has recently attracted the attention of researchers. This technology has become the center of interest of young people because of the real feelings and sensations it creates. People see their environment as 3D because of their eye structure. In this study, it is hypothesized that people lose their perception of depth during sleepy moments and that there is a sudden transition from 3D vision to 2D vision. Regarding these transitions, the EEG signal analysis method was used for deep and comprehensive analysis of 2D and 3D brain signals. In this study, a single-stream anaglyph video of random 2D and 3D segments was prepared. After watching this single video, the obtained EEG recordings were considered for two different analyses: the part involving the critical transition (transition state) and the state analysis of only the 2D versus 3D or 3D versus 2D parts (steady state). The main objective of this study is to see the behavioral changes of brain signals in 2D and 3D transitions. To clarify the impacts of the human brain's power spectral density (PSD) in 2D-to-3D (2D_3D) and 3D-to-2D (3D_2D) transitions of anaglyph video, nine visual healthy individuals were prepared for testing in this pioneering study. Spectrogram graphs based on short time Fourier transform (STFT) were considered to evaluate the power spectrum analysis in each EEG channel of transition or steady state. Thus, in 2D and 3D transition scenarios, important channels representing EEG frequency bands and brain lobes will be identified. To classify the 2D and 3D transitions, the dominant bands and time intervals representing the maximum difference of PSD were selected. Afterward, effective features were selected by applying statistical methods such as standard deviation, maximum (max) and Hjorth parameters to epochs indicating transition intervals. Ultimately, k-nearest neighbors, support vector machine and linear discriminant analysis (LDA) algorithms were applied to classify 2D_3D and 3D_2D transitions. The frontal, temporal and partially parietal lobes show 2D_3D and 3D_2D transitions with a good classification success rate. Overall, it was found that Hjorth parameters and LDA algorithms have 71.11% and 77.78% classification success rates for transition and steady state, respectively.
AtıfScopus6WoS4OpenAlex6İncele - MakaleYalnız künye2022Scopus SJRQ2InSiNet: a deep convolutional approach to skin cancer detection and segmentation
Reis, Hatice Catal; Turk, Veysel; Khoshelham, Kourosh; Kaya, Serhat
Cancer is among the common causes of death around the world. Skin cancer is one of the most lethal types of cancer. Early diagnosis and treatment are vital in skin cancer. In addition to traditional methods, method such as deep learning is frequently used to diagnose and classify the disease. Expert experience plays a major role in diagnosing skin cancer. Therefore, for more reliable results in the diagnosis of skin lesions, deep learning algorithms can help in the correct diagnosis. In this study, we propose InSiNet, a deep learning-based convolutional neural network to detect benign and malignant lesions. The performance of the method is tested on International Skin Imaging Collaboration HAM10000 images (ISIC 2018), ISIC 2019, and ISIC 2020, under the same conditions. The computation time and accuracy comparison analysis was performed between the proposed algorithm and other machine learning techniques (GoogleNet, DenseNet-201, ResNet152V2, EfficientNetB0, RBF-support vector machine, logistic regression, and random forest). The results show that the developed InSiNet architecture outperforms the other methods achieving an accuracy of 94.59%, 91.89%, and 90.54% in ISIC 2018, 2019, and 2020 datasets, respectively. Since the deep learning algorithms eliminate the human factor during diagnosis, they can give reliable results in addition to traditional methods. Graphical abstract: [Figure not available: see fulltext.]
AtıfScopus170WoS313OpenAlex160İncele - MakaleYalnız künye2022Scopus SJRQ3Roza: a new and comprehensive metric for evaluating classification systems
Melek, Mesut; Melek, Negin
Many metrics such as accuracy rate (ACC), area under curve (AUC), Jaccard index (JI), and Cohen’s kappa coefficient are available to measure the success of the system in pattern recognition and machine/deep learning systems. However, the superiority of one system to one other cannot be determined based on the mentioned metrics. This is because such a system can be successful using one metric, but not the other ones. Moreover, such metrics are insufficient when the number of samples in the classes is unequal (imbalanced data). In this case, naturally, by using these metrics, a sensible comparison cannot be made between two given systems. In the present study, the comprehensive, fair, and accurate Roza (Roza means rose in Persian. When different permutations of the metrics used are superimposed in a polygon format, it looks like a flower, so we named it Roza.) metric is introduced for evaluating classification systems. This metric, which facilitates the comparison of systems, expresses the summary of many metrics with a single value. To verify the stability and validity of the metric and to conduct a comprehensive, fair, and accurate comparison between the systems, the Roza metric of the systems tested under the same conditions are calculated and comparisons are made. For this, systems tested with three different strategies on three different datasets are considered. The results show that the performance of the system can be summarized by a single value and the Roza metric can be used in all systems that include classification processes, as a powerful metric.
AtıfScopus7WoS8OpenAlex9İncele - BildiriAçık Erişim2021Automatic Sleep Scoring System Based On Autoregressive Model
Melek, Mesut
Electroencephalography (EEG) signals recorded during sleep are divided into 30-second segments and analyzed by experts and classified into sleep stages, and in this way, diagnosis and treatment of diseases are made. Computer-assisted sleep scoring systems can effectively reduce the burden of experts by classifying sleep stages. In this study, after preprocessing of the EEG signals, four features were obtained for each EEG band (20 features in total) with the autoregressive model and then classified with three different classifiers common in machine learning. The proposed method was tested on the data set called ISRUC sleep (Subgroup 3, ISRUC3) and the results were compared with the results of other studies. The results show that the proposed method is more successful than other existing methods, despite the use of fewer features.
AtıfOpenAlex1İncele - MakaleAçık Erişim2021Scopus SJRQ1Diagnosis of COVID-19 and non-COVID-19 patients by classifying only a single cough sound
Melek, Mesut
In the last month of 2019, a new virus emerged in China, spreading rapidly and affecting the whole world. This virus, which is called corona, is the most contagious type of virus that humanity has ever encountered. The virus has caused a huge crisis worldwide as it leads to severe infections and eventually death in humans. On March 11, 2020, it was announced by the World Health Organization that a COVID-19 outbreak has occurred. Computer-aided digital technologies, which eliminate many problems and provide convenience in people's lives, did not leave humanity alone in this regard and rushed to provide a solution for this unfortunate event. One of the important aspects in which computer-aided digital technologies can be effective is the diagnosis of the disease. Reverse transcription-polymerase chain reaction (RT-PCR), which is a standard and precise technique for diagnosing the disease, is an expensive and time-consuming method. Moreover, its availability is not the same all over the world. For this reason, it can be very attractive and important to distinguish the COVID-19 disease from a cold or flu through a cough sound analysis via smartphones which have entered into the lives of many people in recent years. In this study, we proposed a machine learning-based system to distinguish patients with COVID-19 from non-COVID-19 patients by analyzing only a single cough sound. Two different data sets were used, one accessible for the public and the other available on request. After combining the data sets, the features were obtained from the cough sounds using the mel-frequency cepstral coefficients (MFCCs) method, and then, they were classified with seven different machine learning classifiers. To determine the optimum values of hyperparameters for MFCCs and classifiers, the leave-one-out cross-validation (LOO-CV) strategy was implemented. Based on the results, the k-nearest neighbors classifier based on the Euclidean distance (kNN Euclidean) with the accuracy rate, sensitivity of COVID-19, sensitivity of non-COVID-19, F-measure, and area under the ROC curve (AUC) of 0.9833, 1.0000, 0.9720, 0.9799, and 0.9860, respectively, is more successful than other classifiers. Finally, the best and most effective features were determined for each classifier using the sequential forward selection (SFS) method. According to the results, the proposed system is excellent compared with similar studies in the literature and can be easily used in smartphones and facilitate the diagnosis of COVID-19 patients. In addition, since the used data set includes reflex and unconscious coughs, the results showed that conscious or unconscious coughing has no effect on the diagnosis of COVID-19 patients based on the cough sound.
AtıfScopus38WoS26OpenAlex45İncele - MakaleYalnız künye2020Scopus SJRQ2An EEG-based stereoscopic research of the PSD differences in pre and post 2D&3D movies watching
Manshouri, Negin; Maleki, Masoud; Kayikcioglu, Temel
Despite knowing the reality of three-dimensional (3D) technology in the form of eye fatigue, this technology continues to be retained by people (especially the young community). To check the influences of the human brain's power spectrum density (PSD) pre and post a 2D and 3D movie watching, a five-member test group was arranged. In this study, electroencephalogram (EEG) was used as a neuroimaging method. EEG recordings of five individuals were taken both before and after watching 2D and 3D movies. The main framework of this study was to analyze the effects of power spectrum density (PSD) of the human brain by testing all EEG frequency bands before and after 2D and 3D movie watching, also to determine effective brain lobes, and consequently, EEG channels. After 2D/3D EEG recording, this record was divided into three stages for analysis. These stages consisted of Relax, Watching, and Rest. This benchmarking analysis included I) before and after watching the 2D movie (R2b and R2a), II) before and after watching the 3D movie (R3b and R3a), and III) after watching the 2D/3D movie (R2a and R3a). In the Relax and Rest stages, the 2D/3D EEG power differences in all channels of brain regions for the five EEG bands, including delta (δ), theta (θ), alpha (α), beta (β), and gamma (γ), were analyzed and compared. The PSD based on short-time Fourier transform (STFT) was used to select the dominant bands in this study. Feature extraction was performed on the preprocessed EEG signals using STFT and discrete wavelet transform (DWT). In the 2D analysis, δ, θ, α, and β acted as dominant bands, in 3D, δ, θ, and α were dominant bands, and in 2D/3D, δ and α were selected as meaningful and dominant bands. Partial least-squares regression (PLSR) and support vector machine (SVM) classification algorithms were considered in order to classify the obtained R2a and R3a EEG signals. After the dominant band selection, the correct choice of effective channel combinations in these bands resulted in the percentage of high success of classifiers for Stage III. Taking into account the optimum number of channels, the best percentage of classification accuracy for the average of PLSR and SVM classification results by considering the best channels of SVM and DWT feature extraction method is 91.83% and 80.73% respectively.
AtıfScopus56WoS43İncele