Berkutay Asan

Mr. Berkutay Asan

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

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

Representative Publications

View All Publications
18 December 2024 in Academic Radiology
Verified
Journal Article
Closed access
MSMerve Solak
Merve Solak
MTMurat Tören
Murat Tören
Berkutay AsanBerkutay Asan
Berkutay Asan
EKEsat Kaba
Esat Kaba
MBMehmet Beyazal
Mehmet Beyazal
FÇFatma Beyazal Çeliker
Fatma Beyazal Çeliker

24 July 2025 in Diagnostics
Verified
Journal Article
Open access
PDF file
HEHüseyin Er
Hüseyin Er
MTMurat Tören
Murat Tören
Berkutay AsanBerkutay Asan
Berkutay Asan
EKEsat Kaba
Esat Kaba
MBMehmet Beyazal
Mehmet Beyazal
Objectives: Spinal diseases are commonly encountered health problems with a wide spectrum. In addition to degenerative changes, other common spinal pathologies include metastases and compression fractures. Benign tumors like hemangiomas and infections such as spondylodiscitis are also frequently observed. Although magnetic resonance imaging (MRI) is considered the gold standard in diagnostic imaging, the morphological similarities of lesions can pose significant challenges in differential diagnoses. In recent years, the use of artificial intelligence applications in medical imaging has become increasingly widespread. In this study, we aim to detect and classify vertebral body lesions using the YOLO-v8 (You Only Look Once, version 8) deep learning architecture. Materials and Methods: This study included MRI data from 235 patients with vertebral body lesions. The dataset comprised sagittal T1- and T2-weighted sequences. The diagnostic categories consisted of acute compression fractures, metastases, hemangiomas, atypical hemangiomas, and spondylodiscitis. For automated detection and classification of vertebral lesions, the YOLOv8 deep learning model was employed. Following image standardization and data augmentation, a total of 4179 images were generated. The dataset was randomly split into training (80%) and validation (20%) subsets. Additionally, an independent test set was constructed using MRI images from 54 patients who were not included in the training or validation phases to evaluate the model’s performance. Results: In the test, the YOLOv8 model achieved classification accuracies of 0.84 and 0.85 for T1- and T2-weighted MRI sequences, respectively. Among the diagnostic categories, spondylodiscitis had the highest accuracy in the T1 dataset (0.94), while acute compression fractures were most accurately detected in the T2 dataset (0.93). Hemangiomas exhibited the lowest classification accuracy in both modalities (0.73). The F1 scores were calculated as 0.83 for T1-weighted and 0.82 for T2-weighted sequences at optimal confidence thresholds. The model’s mean average precision (mAP) 0.5 values were 0.82 for T1 and 0.86 for T2 datasets, indicating high precision in lesion detection. Conclusions: The YOLO-v8 deep learning model we used demonstrates effective performance in distinguishing vertebral body metastases from different groups of benign pathologies.

24 April 2025 in Diagnostics
Verified
Journal Article
Open access
PDF file
MBMehmet Beyazal
Mehmet Beyazal
MSMerve Solak
Merve Solak
MTMurat Tören
Murat Tören
Berkutay AsanBerkutay Asan
Berkutay Asan
EKEsat Kaba
Esat Kaba
FÇFatma Beyazal Çeliker
Fatma Beyazal Çeliker
Objectives: Differentiation of hyperdense areas on non-contrast computed tomography (NCCT) images as hemorrhagic transformation (HT) and contrast accumulation (CA) after endovascular thrombectomy (EVT) in acute ischemic stroke (AIS) patients are critical for early antiplatelet and anticoagulant therapy. This study aimed to predict HT and CA on initial NCCT using deep learning. Material and Methods: This study was conducted between January and December 2024. The study included 556 images of 52 patients (21 female and 31 male) who underwent EVT due to AIS, with hyperdense areas observed in the NCCT examination within the first 24 h post-EVT. The evaluated images were labeled as ‘contrast accumulation’ and ‘hemorrhagic transformation’. These labeled images were trained with nine different models under a convolutional neural network (CNN) architecture using a large dataset, such as ImageNet. These models are DenseNet201, InceptionResNet, InceptionV3, NASNetLarge, ResNet50, ResNet101, VGG16, VGG19 and Xception. After training the CNN models, their performance was evaluated using accuracy, loss, validation accuracy, validation loss, F1 score, Receiver Operating Characteristic (ROC) Curve, confusion matrix, confidence interval, and p-value analysis. Results: The models trained in the study were derived from 556 images in data sets obtained from 52 patients; 186 images in training data for CA and 186 images training data for HT (with an increase to 558 images), 115 images used for validation data, and 69 images were compared using test data. In the test set, the Area Under the Curve (AUC) metrics showing sensitivity and specificity values under different cutoff points for the models were as follows: DenseNet201 model AUC = 0.95, InceptionV3 model AUC = 0.93, NasNetLarge model AUC = 0.89, Xception model AUC = 0.91, Inception_ResNet model AUC = 0.84, Resnet50 and Resnet101 models AUC = 0.74. The InceptionV3 model demonstrates the best performance with an F1 score of 0.85. Recall scores generally ranged between 0.62 and 0.85. Conclusions: In our study, hyperdensity areas in initial NCCT images obtained after EVT in AIS patients were successfully differentiated from HT and CA with high accuracy using CNN architectures. Our findings may enable the early identification of patients who would benefit from anticoagulation or antiplatelet therapy to prevent re-occlusion or progression after EVT.

21 September 2024
Verified
Conference Paper
Closed access
MTMurat Tören
Murat Tören
MAMustafa Akiner
Mustafa Akiner
BKBengisu Kaba
Bengisu Kaba
FYFeyza Yazici
Feyza Yazici
Berkutay AsanBerkutay Asan
Berkutay Asan
FBFatih Beriş
Fatih Beriş
The objective of this study is to ascertain the structural variations resulting from the detrimental impact of invertebrates on tea plant leaves. To this end, image processing and deep learning techniques were employed to identify plot regions in Artvin, Rize and Trabzon, where tea cultivation is highly prevalent. The use of drones enabled the acquisition of images in these regions. The results of the examinations indicated the presence of tea pest mites in the region in question. As of recent years, studies were initiated to determine the damage observed in the region, following the commencement of image acquisition. The damage was identified with an accuracy rate of 98% through the utilisation of the YoloV8 algorithm, which represents a sophisticated image processing technology, in conjunction with the convolutional neural network deep learning method. This was achieved through the application of 70 training periods to 800 images. The Edge Step (ST) method was employed to ascertain the geometric differences in the tea leaves, while the Convolutional Neural Network (CNN) was utilised to achieve the minimum classification error. The images were captured using a UAV specifically designed for the study, as it offered a convenient solution even in challenging terrain. The images were subjected to RGB, HSV and black and white colour transformations through the application of image processing techniques. Subsequently, the images were resized and divided into two classes, namely healthy and diseased. The utilisation of these data sets, divided into two classes, facilitated the attainment of more expeditious and precise results through the deployment of the image detection algorithm YOLOv8I-seg. Consequently, the rate of damage propagation was also expressed as a percentage. The study has facilitated the early detection of this damage, which represents a significant threat to tea yield. This has enabled the presentation of valuable data sets that can inform strategies for preventing a decline in yield and enhancing tea quality. These data sets provide an early warning system for combating the harmful organisms that affect the tea plant, which is the region’s primary source of income.

15 August 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
MTMurat Tören
Murat Tören
BKBengisu Kaba
Bengisu Kaba
Berkutay AsanBerkutay Asan
Berkutay Asan
GÖGöktürk Öztürk
Göktürk Öztürk
MAM.MUstafa Akıner
M.MUstafa Akıner
FBFatih Beriş
Fatih Beriş
Visible mite-associated damage in tea plantations is generally assessed through labor-intensive and observer-dependent field inspections. This study developed a hybrid framework combining unpaired CycleGAN image translation with YOLO11x-seg to segment visible mite-associated damage from standard RGB images acquired by unmanned aerial vehicle and close-range cameras. Two CycleGAN models generated NIR-like and RED-like representations, which were combined to produce an NDVI-like vegetation indicator using a threshold of 0.25. The models achieved mean PSNR values of 27.11 dB and 29.11 dB for the NIR-like and RED-like outputs, respectively. YOLO11x-seg achieved a mask AP50 of approximately 0.83 and a maximum F1 score of approximately 0.70. The integrated framework localized and delineated leaf- and canopy-level damage associated with Polyphagotarsonemus latus and Brevipalpus phoenicis, provided vegetation-oriented spatial visualization, and enabled pixel-based quantification of predicted damage coverage. The proposed approach offers a practical basis for more objective automated monitoring and timely management of mite-associated damage in tea plantations.

4 August 2026 in Journal of Imaging Informatics in Medicine
Verified
Journal Article
Closed access
PDF file
FATMA SUMERFATMA SUMER
FATMA SUMER
MTMurat Tören
Murat Tören
Berkutay AsanBerkutay Asan
Berkutay Asan
MSMerve Solak
Merve Solak
NCNurullah Coskuner
Nurullah Coskuner
HOHakan Ozdemir
Hakan Ozdemir