Browsing by Author "Duran, Semra"

Now showing 1 - 3 of 3

Automatic detection of spina bifida occulta with deep learning methods from plain pelvic radiographs
(2023) Maraş, Hadi Hakan; Üreten, Kemal; Maraş, Yüksel; Maraş, Hadi Hakan; Gök, Kevser; Atalar, Ebru; Çayhan, Velihan; 34410
Purpose: Spina bifida occulta (SBO), which is the most common congenital spinal deformity, is often seen in the lower lumbar spine and sacrum. In this study, it is aimed to develop a computer-aided diagnosis method that will help clinicians in the diagnosis of spina bifida occulta from plain pelvic radiographs with deep learning methods and transfer learning method. Materials and methods: The You Only Look Once (YOLO) algorithm was used for object detection, and classification was made by applying transfer learning with a pre-trained VGG-19, ResNet-101, MobileNetV2, and GoogLeNet networks. Our dataset consisted of 206 normal lumbosacral radiographs and 160 SBO lumbosacral radiographs. The performance of the models was evaluated by metrics such as accuracy, sensitivity, specificity, precision, F1 score, and area under the ROC curve (AUC) results. Results: In the detection of SBO, 85.5%, 80.8%, 89.7%, 87.5%, 84%, and 0.92 accuracy, sensitivity, specificity, precision, F1 score, and AUC results were obtained with the pre-trained VGG-19 model, respectively. The pre-trained VGG-19 model performed better than the others. Conclusion: Successful results were obtained in this study performed to the diagnosis of SBO with deep learning methods. A model that will assist physicians in the diagnosis of SBO can be developed with new studies to be conducted with a large number of spinal radiographs.
Deep learning methods in the diagnosis of sacroiliitis from plain pelvic radiographs
(Oxford Univ Press, 2023) Ureten, Kemal; Maras, Yuksel; Duran, Semra; Gok, Kevser
Objectives The aim of this study is to develop a computer-aided diagnosis method to assist physicians in evaluating sacroiliac radiographs. Methods Convolutional neural networks, a deep learning method, were used in this retrospective study. Transfer learning was implemented with pre-trained VGG-16, ResNet-101 and Inception-v3 networks. Normal pelvic radiographs (n = 290) and pelvic radiographs with sacroiliitis (n = 295) were used for the training of networks. Results The training results were evaluated with the criteria of accuracy, sensitivity, specificity and precision calculated from the confusion matrix and AUC (area under the ROC curve) calculated from ROC (receiver operating characteristic) curve. Pre-trained VGG-16 model revealed accuracy, sensitivity, specificity, precision and AUC figures of 89.9%, 90.9%, 88.9%, 88.9% and 0.96 with test images, respectively. These results were 84.3%, 91.9%, 78.8%, 75.6 and 0.92 with pre-trained ResNet-101, and 82.0%, 79.6%, 85.0%, 86.7% and 0.90 with pre-trained inception-v3, respectively. Conclusions Successful results were obtained with all three models in this study where transfer learning was applied with pre-trained VGG-16, ResNet-101 and Inception-v3 networks. This method can assist clinicians in the diagnosis of sacroiliitis, provide them with a second objective interpretation and also reduce the need for advanced imaging methods such as magnetic resonance imaging.
Diagnosis of osteoarthritic changes, loss of cervical lordosis, and disc space narrowing on cervical radiographs with deep learning methods
(Turkish Joint Diseases Foundation, 2022) Maras, Yuksel; Tokdemir, Gül; Tokdemir, Gul; Ureten, Kemal; Atalar, Ebru; Duran, Semra; Maras, Hakan; 17411; 34410
Objectives: In this study, we aimed to differentiate normal cervical graphs and graphs of diseases that cause mechanical neck pain by using deep convolutional neural networks (DCNN) technology. Materials and methods: In this retrospective study, the convolutional neural networks were used and transfer learning method was applied with the pre-trained VGG-16, VGG-19, Resnet-101, and DenseNet-201 networks. Our data set consisted of 161 normal lateral cervical radiographs and 170 lateral cervical radiographs with osteoarthritis and cervical degenerative disc disease. Results: We compared the performances of the classification models in terms of performance metrics such as accuracy,