WoS İndeksli Yayınlar Koleksiyonu
Permanent URI for this collectionhttps://hdl.handle.net/20.500.12573/394
Browse
5 results
Search Results
Article Citation - WoS: 27Citation - Scopus: 30Super Resolution Convolutional Neural Network Based Pre-Processing for Automatic Polyp Detection in Colonoscopy Images(Pergamon-Elsevier Science Ltd, 2021-03) Tas, Merve; Yilmaz, BulentColonoscopy is the most common methodology used to detect polyps on the colon surface. Increasing the image resolution has the potential to improve the automatic colonoscopy based diagnosis and polyp detection and localization. In this study, we proposed a pre-processing approach that uses convolutional neural network based super resolution method (SRCNN) to increase the resolution of the training colonoscopy images before the localization of polyps. We also investigated the use of CNN based models such as the Single Shot MultiBox Detector (SSD) and Faster Regional CNN (RCNN) for real-time polyp detection and localization. Our results showed that using SRCNN method before the training process provides better results in terms of accuracy in both models compared to the low-resolution cases. Furthermore, we reached an F2 score of 0.945 for the correct localization of colon polyps using Faster RCNN with ResNet-101 feature extractor.Article Citation - WoS: 26Citation - Scopus: 48Metabolic Imaging Based Sub-Classification of Lung Cancer(IEEE-Inst Electrical Electronics Engineers Inc, 2020) Bicakci, Mustafa; Ayyildiz, Oguzhan; Aydin, Zafer; Basturk, Alper; Karacavus, Seyhan; Yilmaz, BulentLung cancer is one of the deadliest cancer types whose 84% is non-small cell lung cancer (NSCLC). In this study, deep learning-based classification methods were investigated comprehensively to differentiate two subtypes of NSCLC, namely adenocarcinoma (ADC) and squamous cell carcinoma (SqCC). The study used 1457 F-18-FDG PET images/slices with tumor from 94 patients (88 men), 38 of which were ADC and the rest were SqCC. Three experiments were carried out to examine the contribution of peritumoral areas in PET images on subtype classification of tumors. We assessed multilayer perceptron (MLP) and three convolutional neural network (CNN) models such as SqueezeNet, VGG16 and VGG19 using three kinds of images in these experiments: 1) Whole slices without cropping or segmentation, 2) cropped image portions (square subimages) that include the tumor and 3) segmented image portions corresponding to tumors using random walk method. Several optimizers and regularization methods were used to optimize each model for the diagnostic classification. The classification models were trained and evaluated by performing stratified 10-fold cross validation, and F-score and area-under-curve (AUC) metrics were used to quantify the performance. According to our results, it is possible to say that inclusion of peritumoral regions/tissues both contributes to the success of models and makes segmentation effort unnecessary. To the best of our knowledge, deep learning-based models have not been applied to the subtype classification of NSCLC in PET imaging, therefore, this study is a significant cornerstone providing thorough comparisons and evaluations of several deep learning models on metabolic imaging for lung cancer. Even simpler deep learning models are found promising in this domain, indicating that any improvement in deep learning models in machine learning community can be reflected well in this domain as well.Article Citation - WoS: 20Citation - Scopus: 26Medical Infrared Thermal Image Based Fatty Liver Classification Using Machine and Deep Learning(Taylor & Francis Ltd, 2023-01-10) Ozdil, Ahmet; Yilmaz, BulentNon-alcoholic fatty liver disease (NAFLD) causes accumulation of excess fat in the liver affecting people who drink little to no alcohol. Non-alcoholic steatohepatitis (NASH) is an aggressive form of fatty liver disease (inflammation in the liver), may progress to cirrhosis and liver failure. Liver function tests, ultrasound (US) and magnetic resonance imaging (MRI) are used to help diagnose and monitor liver disease or damage. In this study, the feasibility of medical infrared thermal imaging (MITI) in automatic detection of NAFLD was investigated, and 167 MITI images (44 positive) from 32 patients (7 positive) were evaluated using image processing and classification methods. Convolutional neural network (CNN) architectures and texture analysis methods were used in the feature selection phase. After feature selection and binary classification, the highest values from different setups for recall, f-score, specificity, accuracy, and area-under-curve (AUC) were 1.00, 1.00, 0.83, 1.0, 0.94, and 0.92, respectively. The highest values were achieved by CNN based methods on different datasets, however, texture analysis method performed lower. Here, it is shown that some of the CNN architectures have high potential on extracting features from thermal images. Finally, machine and deep learning approaches can be combined in detecting NAFLD using infrared thermal images.Article Citation - WoS: 13Citation - Scopus: 15Comparative Analysis of Dimensionality Reduction Techniques for Cybersecurity in the SwaT Dataset(Springer, 2023-07-08) Bozdal, Mehmet; Ileri, Kadir; Ozkahraman, AliThe Internet of Things (IoT) has revolutionized the functionality and efficiency of distributed cyber-physical systems, such as city-wide water treatment systems. However, the increased connectivity also exposes these systems to cybersecurity threats. This research presents a novel approach for securing the Secure Water Treatment (SWaT) dataset using a 1D Convolutional Neural Network (CNN) model enhanced with a Gated Recurrent Unit (GRU). The proposed method outperforms existing methods by achieving 99.68% accuracy and an F1 score of 98.69%. Additionally, the paper explores dimensionality reduction methods, including Autoencoders, Generalized Eigenvalue Decomposition (GED), and Principal Component Analysis (PCA). The research findings highlight the importance of balancing dimensionality reduction with the need for accurate intrusion detection. It is found that PCA provided better performance compared to the other techniques, as reducing the input dimension by 90.2% resulted in only a 2.8% and 2.6% decrease in the accuracy and F1 score, respectively. This study contributes to the field by addressing the critical need for robust cybersecurity measures in IoT-enabled water treatment systems, while also considering the practical trade-off between dimensionality reduction and intrusion detection accuracy.Article Citation - WoS: 17Citation - Scopus: 19A Deep Ensemble Approach for Long-Term Traffic Flow Prediction(Springer Heidelberg, 2024-01-27) Cini, Nevin; Aydin, ZaferIn the last 50 years, with the growth of cities and increase in the number of vehicles and mobility, traffic has become troublesome. As a result, traffic flow prediction started to attract attention as an important research area. However, despite the extensive literature, traffic flow prediction still remains as an open research problem, specifically for long-term traffic flow prediction. Compared to the models developed for short-term traffic flow prediction, the number of models developed for long-term traffic flow prediction is very few. Based on this shortcoming, in this study, we focus on long-term traffic flow prediction and propose a novel deep ensemble model (DEM). In order to build this ensemble model, first, we developed a convolutional neural network (CNN), a long short-term memory (LSTM) network and a gated recurrent unit (GRU) network as deep learning models, which formed the base learners. In the next step, we combine the output of these models according to their individual forecasting success. We use another deep learning model to determine the success of the individual models. Our proposed model is a flexible ensemble prediction model that can be updated based on traffic data. To evaluate the performance of the proposed model, we use a publicly available dataset. Experimental results show that the developed DEM model has a mean square error of 0.06 and a mean absolute error of 0.15 for single-step prediction; it shows that achieves a mean square error of 0.25 and a mean absolute error of 0.32 for multi-step prediction. We compared our proposed model with many models in different categories; individual deep learning models (i.e., LSTM, CNN, GRU), selected traditional machine learning models (i.e., linear regression, decision tree regression, k-nearest-neighbors regression) and other ensemble models such as random-forest regression. These results also support the claim that ensemble learning models perform better than individual models.