Sürdürülebilir Kalkınma Amaçları
Yazarlar
Yayınlar
- Yüksek Lisans TeziAçık Erişim2025Otomatik avuç içi bölütlenmesinin gerçekleştirilmesinde farklı derin öğrenme modellerinin performanslarının incelenmesi
Kadir Yalçın
Tıbbi görüntüleme teknolojilerindeki ilerlemeler, özellikle derin öğrenme tabanlı yöntemlerin kullanımıyla, el bölütleme gibi spesifik görevlerde dikkat çekici başarılar elde etmiştir. El bölütleme, biyometrik doğrulama sistemlerinde kimlik tespiti için kritik bir öneme sahiptir. Avuç içi damar yapıları, parmak uzunlukları ve deri dokusu gibi benzersiz biyometrik özelliklerin doğru bir şekilde çıkarılması, el bölütleme algoritmalarının doğruluğuna bağlıdır. Bu tez çalışmasında, el bölütleme probleminin çözümü için oto kodlayıcı tabanlı U-Net, MA-Net, FPN ve PSP-Net modelleri, Vision Transformer, EfficientNet-B5 ve ResNet-34 omurgaları ile değerlendirilmiştir. Modeller, Doğruluk (ACC), Intersection over Union (IoU), Dice katsayısı ve F1 skor gibi metriklerle analiz edilmiştir. Sonuçlar, Vision Transformer ile entegre edilen U-Net modelinin bölütleme de en yüksek başarı sağladığını göstermiştir (IoU: 0.8917, Dice: 0.9892). MA-Net, ResNet-34 omurgasıyla kullanıldığında en yüksek doğruluk değerine (0.9921) ulaşmıştır. Ayrıca bu kombinasyon, en yüksek F1 skorunu (0.9584) da elde etmiştir. EfficientNet-B5, MA-Net ve FPN ile etkili sonuçlar sunmuştur. Çalışmanın sonuçları, derin öğrenme modelleriyle elde edilen bölütleme başarısının biyometrik doğrulama sistemlerinin güvenilirliğini artırabileceğini ortaya koymaktadır. Özellikle U-Net - Vision Transformer ve MA-Net – ResNet-34 kombinasyonları, elin bölütlenmesinde yüksek performans göstermiştir. Bu bulgular, el bölütleme uygulamalarında model ve omurga seçimlerinin kritik olduğunu vurgulamaktadır.
- MakaleAçık Erişim2025Scopus SJRQ1A multi-stage fusion deep learning framework merging local patterns with attention-driven contextual dependencies for cancer detection
Catal Reis, Hatice; Turk, Veysel
Cancer is a severe threat to public health. Early diagnosis of disease is critical, but the lack of experts in this field, the personal assessment process, the clinical workload, and the high level of similarity in disease classes make it difficult. In recent years, deep learning-based artificial intelligence models have shown promise, with the potential to increase diagnosis speed and accuracy. These models attract attention with their automatic learning and adaptation capabilities. In this study, the deep learning-based PADBSRNet model and the PADBSRNet-Vision Transformer (ViT) hybrid method are proposed for the detection of brain tumors and skin and lung cancers. PADBSRNet is a comprehensive deep neural network architecture that integrates separable and traditional convolution layers, multiple attention mechanisms, bidirectional recurrent neural networks, and cross-connections/multi-stage feature fusion strategies. This architecture offers significant advantages in terms of effectively extracting local-global, contextual features and accurately modeling long-term dependencies in image classification tasks. The second proposed approach developed a hybrid method that combines the advantages of the PADBSRNet model and the ViT model. Experimental analysis on medical datasets such as the Figshare Brain Tumor Dataset, IQ-OTH/NCCD Dataset, and Skin Cancer: Malignant vs. Benign Dataset has evaluated the proposed models' performances. According to the experimental results, the PADBSRNet model has shown successful performance with 95.24 %, 99.55 %, and 88.61 % accuracy rates, respectively. The experimental findings show that the proposed deep learning model can effectively learn the complex relationships and hidden patterns of cancer disease, thus producing applicable and effective results in cancer diagnosis.
AtıfScopus3OpenAlex2İncele - MakaleYalnız künye2021Scopus SJRQ1Cost-oriented LSTM methods for possible expansion of control charting signals
Ünlü, Ramazan
Manual quality control may result in delayed detection of a system defect, or none at all, potentially resulting in malfunctions that can lead to disruption of the system, incur extra costs, or complete downtime of the system. To mitigate such issues, more advanced statistics and machine learning-based systems such as SVM, WSVM, etc. are used to automatically detect error signals during the process. However, these methodologies are not developed for sequential dataset such as a classical CCPR dataset. As a result, in this study, we have implemented a preliminary Cost-Oriented Long-Short Term Memory (LSTM), which is designed to learn from a sequential dataset and compare it with SVM and WSVM which are traditional methods utilized in the field. Additionally, we compared the performance of methods in terms of early detection of an abnormal pattern. Based on the findings, the Cost-Oriented LSTM method outperforms SVM and WSVM in the majority of abnormal patterns in terms of both classification and early detection performance. Global accuracy score which is the average of accuracy rates in all combination of abnormal parameters and window lengths shows that LSTM gives a better accuracy score than SVM and WSVM in all seven abnormal patterns detection.
AtıfScopus22WoS19OpenAlex22İncele - Kitap BölümüYalnız künye2020A novel similarity measure for single-valued neutrosophic sets and their applications in medical diagnosis, taxonomy, and clustering analysis
Şahin, Ridvan; Karabacak, Mesut
As a generation of Atanassov's intuitionistic fuzzy set, the single-valued neutrosophic set is an important modeling approach for expressing and processing the inconsistent and indeterminate information very well. Similarity measure is an important tool frequently used in a variety of areas, from clustering analysis to medical diagnosis. Although neutrosophic literature has many similarity measures, they have some disadvantages that do not provide the general conditions of the similarity measure for some specific values. In this chapter, we present a novel similarity measure to handle the relationship between two single-valued neutrosophic sets. It uses a matrix norm and a strictly increasing (or decreasing) binary function called fuzzy implication, and provides axiomatic definition properties of similarity measure. The advantage of using fuzzy implications is that it offers not only different final options to decision-makers but also gives a parameterized class of similarity measures of SVNSs. It is appeared that the developed similarity measure gives better results when compared to other existing similarity measures. Finally, several numerical examples related to pattern recognition, such as medical diagnosis and taxonomy approach, and clustering analysis are performed to demonstrate the practical applicability of the proposed similarity measure.
AtıfScopus17OpenAlex14İncele
