Nurullah Düger

Dr. Nurullah Düger

orcid

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
5 March 2026 in Diagnostics
Verified
Journal Article
Open access
PDF file
NDNurullah Düger
Nurullah Düger
FTFurkan Talo
Furkan Talo
GTGÜLÜÇAĞ GİRAY TEKİN
GÜLÜÇAĞ GİRAY TEKİN
BDBurak Dağtekin
Burak Dağtekin
Mucahit KaradumanMucahit Karaduman
Mucahit Karaduman
TYTuba Talo Yıldırım
Tuba Talo Yıldırım
Objectives: This study aimed to develop and validate a hybrid deep learning model combining Convolutional Neural Networks (CNN) and Vision Transformers (ViT) to automatically classify maxillary sinus membrane morphologies on Cone-Beam Computed Tomography (CBCT) images, distinguishing between Normal, Flat, Polypoid, and Obstruction types. Methods: A dataset of 959 CBCT images was collected and categorized into four morphological classes: Normal, Flat, Polypoid and Obstruction. A custom hybrid model was developed, integrating a lightweight residual CNN for local feature extraction, learnable weighted feature fusion with a bidirectional feature pyramid network and a Transformer encoder for global context modeling. The performance of proposed model was compared against six different architectures, including ResNet50, MobileNetV3L and standard ViT models, using accuracy, precision, recall and F1-score metrics. Results: The proposed hybrid model achieved the highest overall accuracy of 98.44%, outperforming six strong CNN and ViT models including ResNet50 (97.92%) and ViT-B16 (86.46%) models. In class-wise analysis, the model demonstrated superior diagnostic capability, particularly for the “Obstruction” class, achieving 100% accuracy. High discrimination was also observed for “Flat” (98.21%) and “Polypoid” (98.04%) morphologies, confirming the model’s sensitivity to shape-based features. Conclusions: The proposed hybrid CNN-ViT model successfully classifies maxillary sinus membrane morphologies with high accuracy, effectively overcoming the limitations of standard ViT models on limited datasets. Detection of membrane morphology is vital for predicting surgical risks like membrane perforation and post-operative sinusitis. This model serves as a reliable clinical decision support tool, enabling clinicians to objectively assess specific risk factors before implant surgery and sinus floor elevation.

21 January 2026 in Biomedicines
Verified
Journal Article
Open access
PDF file
SASuna Aydin
Suna Aydin
İPİsmail Polat
İsmail Polat
KTKevser Tural
Kevser Tural
NDNurullah Düger
Nurullah Düger
KUKader Ugur
Kader Ugur
DLDo-Youn Lee
Do-Youn Lee
Background/Objectives: The underlying molecular mechanisms of deep vein thrombosis (DVT), which continues to be a major global public health concern, remain unclear. A key component of anticoagulant therapy, heparin (HP) interacts with heparin-binding growth factors including pleiotrophin (PTN) and midkine (MK), both of which have basic amino acid-rich domains that have a strong affinity for HP. The purpose of this study was to determine if changes in the levels of circulating HP, MK, and PTN are linked to the onset of acute DVT. Methods: Thirty patients diagnosed with acute DVT by venous Doppler ultrasonography (VDU) and 28 healthy controls with normal VDU findings were enrolled. Serum HP, MK, and PTN concentrations were measured using ELISA. In DVT patients, blood samples were obtained before and after routine subcutaneous low-molecular-weight heparin treatment; controls provided a single blood sample. ROC curve analysis was used to assess diagnostic performance. Results: Prior to treatment, patients with acute DVT exhibited significantly lower serum HP levels (p < 0.05) and significantly higher MK and PTN levels compared with healthy controls (both p < 0.05). Following heparin administration, serum HP levels increased significantly (p < 0.05), while MK and PTN levels showed a decreasing trend that did not reach statistical significance (p > 0.05). ROC curve analysis demonstrated limited diagnostic performance for HP (sensitivity 10.3%, specificity 68.8%), PTN (62.1%, 54.2%), and MK (82.8%, 35.4%). Conclusions: Decreased circulating HP and increased MK and PTN levels are characteristics of acute DVT that may indicate endogenous HP sequestration through binding to these growth factors. This imbalance could lead to less free HP being available, which would encourage the formation of thrombus. Therapeutic approaches that target MK- and PTN-mediated HP interactions may constitute a unique approach for the therapy of acute DVT, as evidenced by the partial normalization seen after exogenous heparin delivery.

10 March 2026 in Journal of Oral Implantology
Verified
Journal Article
Closed access
AKAyşe Rençber Kızılkaya
Ayşe Rençber Kızılkaya
Mucahit KaradumanMucahit Karaduman
Mucahit Karaduman
HKHarun Kurt
Harun Kurt
TYTuba Yildirim
Tuba Yildirim
NDNurullah Düger
Nurullah Düger
MYMuhammed YILDIRIM
Muhammed YILDIRIM
Dental implants are commonly used to restore the function and esthetics of missing teeth, and their long-term success depends on appropriate maintenance and follow-up interventions. One of the major clinical challenges is clinicians’ inability to identify the brand of a patient’s previously placed implants, which complicates the selection of compatible prosthetic components and instruments for subsequent treatments. This study aimed to develop a deep learning–based approach to identify dental implant brands automatically. A hybrid model based on Vision Transformer (ViT) architectures was proposed. To enhance efficiency, feature maps from three state-of-the-art ViT models were integrated and then classified with multiple classifiers. The proposed model achieved an accuracy of 87.92%, outperforming widely used benchmark methods in the literature. These findings demonstrate the proposed model's potential to help clinicians accurately identify implant brands, thereby facilitating appropriate prosthetic procedures and improving clinical workflow.

29 December 2023 in Journal of Oral Health and Oral Epidemiology
Verified
Journal Article
Open access
PDF file
DADicle Altindal
Dicle Altindal
FKFiliz Acun Kaya
Filiz Acun Kaya
TYTuba Yildirim
Tuba Yildirim
EUErsin Uysal
Ersin Uysal
NDNurullah Düger
Nurullah Düger
Background: The aim of this study was to examine the general behavioral status of individuals who applied to a periodontology clinic during the COVID-19 pandemic to evaluate their health anxiety levels and to compare their anxiety score averages in terms of various factors. Methods: This study was carried out using the face-to-face survey method. Participants were asked to fill out a demographic data form, a survey form asking about pandemic-related behaviors, and the Short Health Anxiety Inventory (SHAI). The effects of factors such as age, gender, marital status, educational status, working in a health institution, and having had COVID-19 disease on health anxiety levels were investigated. The Kolmogorov-Smirnov test, Shapiro-Wilk test, independent samples t test, ANOVA, and Tukey’s HSD test were used for statistical analysis (P<0.05). Results: A total of 400 volunteer participants were included (252 females and 148 males). Anxiety score averages showed a statistically significant difference in terms of gender and education (P=0.023 and P=0.001, respectively). Accordingly, the females’ mean SHAI score (17.45±7.053) was found to be higher than the males’ (15.69±8.161). The mean SHAI scores of primary school graduates (20.86±6.105) were higher than those of high school (15.95±8.197) and university graduates (16.11±6.367). There was no significant difference in terms of marital status, working in a health institution, and being diagnosed with COVID-19. Also, 48.2% of the participants only used surgical masks. Conclusion: Most participants declared that their confidence in healthcare workers increased during the epidemic. Females and individuals with low educational status had higher health anxiety in this study.

25 May 2026 in Diagnostics
Verified
Journal Article
Open access
PDF file
TYTuba Talo Yıldırım
Tuba Talo Yıldırım
ADAybike Cengiz Dağtekin
Aybike Cengiz Dağtekin
NDNurullah Düger
Nurullah Düger
AKAyşe Rençber Kızılkaya
Ayşe Rençber Kızılkaya
FTFurkan Talo
Furkan Talo
MYMuhammed YILDIRIM
Muhammed YILDIRIM
Background/Objectives: The aim of this study is to test different convolutional neural network (CNN) and Transformer-based models to detect and classify vertical misfit at the abutment-prosthesis interface on panoramic radiographs, and to develop a hybrid deep learning model enhanced with attention mechanisms. Methods: A dataset consisting of a total of 566 images, manually classified as 249 ‘fit’ and 317 ‘misfit’ cases by two experts, was created. Images were resized to 224 × 224 and divided into training, validation, and test groups. The deep learning model yielding the most successful results was determined as the backbone; a hybrid model was developed by integrating three different attention modules (SE, CBAM, and ECA) into this structure. Model performance was evaluated using accuracy, precision, sensitivity, and F1 score metrics. Results: CNN-based models (RegNetY-800MF, ConvNeXt-Tiny, EfficientNetV2-S, ResNet50) performed better than Transformer-based models (DeiT, Swin-Tiny) in all metrics. The proposed hybrid model exhibited the highest success among all tested models with a 99.12% accuracy rate. This model reached a 100% precision value in the misfit group and yielded no false positive results. The F1 scores of the hybrid model were recorded as 99.01% for the fit group and 99.21% for the misfit group. Conclusions: The findings of this study demonstrate that attention-enhancing deep learning frameworks have the potential to significantly improve the diagnostic utility of routine panoramic radiographs. It shows that panoramic imaging, when supported by advanced artificial intelligence, can provide valuable diagnostic support in detecting vertical misfit. The developed model has the potential to become a reliable clinical decision support system.

6 November 2026 in Journal of Clinical Medicine
Verified
Journal Article
Open access
PDF file
NDNurullah Düger
Nurullah Düger
BDBurak Dağtekin
Burak Dağtekin
FTFurkan Talo
Furkan Talo
MAMelek Atille Aydin
Melek Atille Aydin
Mucahit KaradumanMucahit Karaduman
Mucahit Karaduman
Tolga TozumTolga Tozum
Tolga Tozum
Background/Objectives: Early diagnosis of periodontal disease is critical to prevent tissue destruction and tooth loss. This study evaluates six modern deep learning architectures for four-class radiographic periodontal bone loss grading on panoramic radiographs. Methods: This retrospective study analyzed 1053 images. Two periodontologists independently classified each image according to the most severely affected tooth into four bone loss categories: Healthy, Mild (50%). Six convolutional, transformer, and hybrid backbones (ConvNeXt-Tiny, MaxViT-Tiny, EfficientViT-B1, FocalNet-Tiny, CAFormer-S18, and DaViT-Tiny) were trained on a 70/10/20 stratified split using an NVIDIA Tesla T4 GPU, the AdamW optimizer, Cosine Annealing, and early stopping. Performance was assessed using standard classification metrics, Matthews Correlation Coefficient, Cohen’s and Quadratic Weighted Kappa, Expected Calibration Error, Area Under the Curve (AUC), McNemar’s test, and bootstrap 95% confidence intervals. Results: FocalNet-Tiny achieved the highest numerical scores among the tested models, reaching 83.41% accuracy, 84.70% balanced accuracy, 83.50% macro-F1, and 0.9610 OvR AUC. The p-values obtained from the comparisons of the models were all above 0.05, indicating that the classification performances of the evaluated models are quite similar. Bootstrap analysis confirmed stable prediction metrics for the top-scoring model. Conclusions: This study provides a controlled benchmark of modern deep learning backbones for four-class radiographic periodontal bone loss grading. FocalNet-Tiny achieved the highest numerical metrics and stable bootstrap estimates, yet performed comparably to other contemporary architectures such as ConvNeXt-Tiny. Direct radiographic classification shows promise for automated radiographic pre-screening; however, multi-center external validation and integration with clinical periodontal parameters are required before clinical implementation.