Scopus İndeksli Yayınlar Koleksiyonu
Permanent URI for this collectionhttps://hdl.handle.net/11727/4809
Browse
5 results
Search Results
Item Use Of Deep Learning Methods For Hand Fracture Detection From Plain Hand Radiographs(2022) Ureten, Kemal; Sevinc, Huseyin Fatih; Igdeli, Ufuk; Onay, Aslihan; Maras, Yuksel; https://orcid.org/0000-0003-4213-9126; 35099027BACKGROUND: Patients with hand trauma are usually examined in emergency departments of hospitals. Hand fractures are frequently observed in patients with hand trauma. Here, we aim to develop a computer-aided diagnosis (CAD) method to assist physicians in the diagnosis of hand fractures using deep learning methods. METHODS: In this study, Convolutional Neural Networks (CNN) were used and the transfer learning method was applied. There were 275 fractured wrists, 257 fractured phalanx, and 270 normal hand radiographs in the raw dataset. CNN, a deep learning method, were used in this study. In order to increase the performance of the model, transfer learning was applied with the pre-trained VGG-16, GoogLeNet, and ResNet-50 networks. RESULTS: The accuracy, sensitivity, specificity, and precision results in Group 1 (wrist fracture and normal hand) dataset were 93.3%, 96.8%, 90.3%, and 89.7% , respectively, with VGG-16, were 88.9%, 94.9%, 84.2%, and 82.4%, respectively, with Resnet-50, and were 88.1%, 90.6%, 85.9%, and 85.3%, respectively, with GoogLeNet. The accuracy, sensitivity, specificity, and precision results in Group 2 (phalanx fracture and normal hand) dataset were 84.0%, 84.1%, 83.8%, and 82.8%, respectively, with VGG-16, were 79.4%, 78.5%, 80.3%, and 79.7%, respectively, with Resnet-50, and were 81.7%, 81.3%, 82.1%, and 81.3%, respectively, with GoogLeNet. CONCLUSION: We achieved promising results in this CAD method, which we developed by applying methods such as transfer learning, data augmentation, which are state-of-the-art practices in deep learning applications. This CAD method can assist physicians working in the emergency departments of small hospitals when interpreting hand radiographs, especially when it is difficult to reach qualified colleagues, such as night shifts and weekends.Item A deep learning approach for sepsis monitoring via severity score estimation(2021) Asuroglu, Tunc; Ogul, Hasan; 33157471Background and objective: Sepsis occurs in response to an infection in the body and can progress to a fatal stage. Detection and monitoring of sepsis require multi-step analysis, which is time-consuming, costly and requires medically trained personnel. A metric called Sequential Organ Failure Assessment (SOFA) score is used to determine the severity of sepsis. This score depends heavily on laboratory measurements. In this study, we offer a computational solution for quantitatively monitoring sepsis symptoms and organ systems state without laboratory test. To this end, we propose to employ a regression-based analysis by using only seven vital signs that can be acquired from bedside in Intensive Care Unit (ICU) to predict the exact value of SOFA score of patients before sepsis occurrence. Methods: A model called Deep SOFA-Sepsis Prediction Algorithm (DSPA) is introduced. In this model, we combined Convolutional Neural Networks (CNN) features with Random Forest (RF) algorithm to predict SOFA scores of sepsis patients. A subset of Medical Information Mart in Intensive Care (MIMIC) III dataset is used in experiments. 5154 samples are extracted as input. Ten-fold cross validation test are carried out for experiments. Results: We demonstrated that our model has achieved a Correlation Coefficient (CC) of 0.863, a Mean Absolute Error (MAE) of 0.659, a Root Mean Square Error (RMSE) of 1.23 for predictions at sepsis onset. The accuracies of SOFA score predictions for 6 hours before sepsis onset were 0.842, 0.697, and 1.308, in terms of CC, MAE and RMSE, respectively. Our model outperformed traditional machine learning and deep learning models in regression analysis. We also evaluated our model's prediction performance for identifying sepsis patients in a binary classification setup. Our model achieved up to 0.982 AUC (Area Under Curve) for sepsis onset and 0.972 AUC for 6 hours before sepsis, which are higher than those reported by previous studies. Conclusions: By utilizing SOFA scores, our framework facilitates the prognose of sepsis and infected organ systems state. While previous studies focused only on predicting presence of sepsis, our model aims at providing a prognosis solution for sepsis. SOFA score estimation process in ICU depends on laboratory environment. This dependence causes delays in treating patients, which in turn may increase the risk of complications. By using easily accessible non-invasive vital signs that are routinely collected in ICU, our framework can eliminate this delay. We believe that the estimation of the SOFA score will also help health professionals to monitor organ states. (C) 2020 Elsevier B.V. All rights reserved.Item Classification of Canine Maturity and Bone Fracture Time Based on X-Ray Images of Long Bones(2021) Ergun, Gulnur Begum; Guney, Selda; 0000-0002-0573-1326; 0000-0001-8469-5484Veterinarians use X-rays for almost all examinations of clinical fractures to determine the appropriate treatment. Before treatment, vets need to know the date of the injury, type of the broken bone, and age of the dog. The maturity of the dog and the time of the fracture affects the approach to the fracture site, the surgical procedure and needed materials. This comprehensive study has three main goals: determining the maturity of the dogs (Task 1), dating fractures (Task 2), and finally, detecting fractures of the long bones in dogs (Task 3). The most popular deep neural networks are used: AlexNet, ResNet-50 and GoogLeNet. One of the most popular machine learning algorithms, support vector machines (SVM), is used for comparison. The performance of all sub-studies is evaluated using accuracy and F1 score. Each task has been successful with different network architecture. ResNet-50, AlexNet and GoogLeNet are the most successful algorithms for the three tasks, with F1 scores of 0.75, 0.80 and 0.88, respectively. Data augmentation is performed to make models more robust, and the F1 scores of the three tasks were 0.80, 0.81, and 0.89 using ResNet-50, which is the most successful model. This preliminary work can be developed into support tools for practicing veterinarians that will make a difference in the treatment of dogs with fractured bones. Considering the lack of work in this interdisciplinary field, this paper may lead to future studies.Item Femoral neck fracture detection in X-ray images using deep learning and genetic algorithm approaches(2020) Beyaz, Salih; Acici, Koray; Sumer, Emre; 0000-0002-5788-5116; 32584712; K-8820-2019Objectives: This study aims to detect frontal pelvic radiograph femoral neck fracture using deep learning techniques. Patients and methods: This retrospective study was conducted between January 2013 and January 2018. A total of 234 frontal pelvic X-ray images collected from 65 patients (32 males, 33 females; mean age 74.9 years; range, 33 to 89 years) were augmented to 2106 images to achieve a satisfactory dataset. A total of 1,341 images were fractured femoral necks while 765 were non-fractured ones. The proposed convolutional neural network (CNN) architecture contained five blocks, each containing a convolutional layer, batch normalization layer, rectified linear unit, and maximum pooling layer. After the last block, a dropout layer existed with a probability of 0.5. The last three layers of the architecture were a fully connected layer of two classes, a softmax layer and a classification layer that computes cross entropy loss. The training process was terminated after 50 epochs and an Adam Optimizer was used. Learning rate was dropped by a factor of 0.5 on every five epochs. To reduce overfitting, regularization term was added to the weights of the loss function. The training process was repeated for pixel sizes 50x50, 100x100, 200x200, and 400x400. The genetic algorithm (GA) approach was employed to optimize the hyperparameters of the CNN architecture and to minimize the error after testing the model created by the CNN architecture in the training phase. Results: Performance in terms of sensitivity, specificity, accuracy, F1 score, and Cohen's kappa coefficient were evaluated using five-fold cross validation tests. Best performance was obtained when cropped images were rescaled to 50x50 pixels. The kappa metric showed more reliable classifier performance when 50x50 pixels image size was used to feed the CNN. The classifier performance was more reliable according to other image sizes. Sensitivity and specificity rates were computed to be 83% and 73%, respectively. With the inclusion of the GA, this rate increased by 1.6%. The detection rate of fractured bones was found to be 83%. A kappa coefficient of 55% was obtained, indicating an acceptable agreement. Conclusion: This experimental study utilized deep learning techniques in the detection of bone fractures in radiography. Although the dataset was unbalanced, the results can be considered promising. It was observed that use of smaller image size decreases computational cost and provides better results according to evaluation metrics.Item Detection of Basic Human Physical Activities With Indoor Outdoor Information Using Sigma-Based Features and Deep Learning(2019) Memis, Gokhan; Sert, Mustafa; 0000-0002-5758-4321; 0000-0002-7056-4245; AAB-8673-2019The devices created on account of the developments in wearable technology are increasingly becoming a part of our daily lives. In particular, sensors have enhanced the usefulness of such devices. The aim of this paper is to detect human physical activity along with indoor/outdoor information by using mobile phones and a separate oxygen saturation sensor. There is no relevant dataset in the literature for this type of detection. For this purpose, data from four different types of human physical activity was collected through mobile phone and oxygen saturation sensors; 12 people aged between 20-65 years participated in the study. During the data collection process, different physical activities under different environmental conditions were performed by the subjects in 10 min. As a next step, a novel deep neural network (DNN) model specifically designed for physical activity recognition was proposed. In order to improve accuracy and reduce the computational complexity, standard deviation (sigma)-based features were introduced. To evaluate its efficacy, we conducted comparisons with selected machine learning algorithms on our proposed dataset. The results on our dataset indicate that the multimodal sigma-based features give the best classification accuracy of 81.60% using our proposed DNN method. Furthermore, the accuracy of the classification made with our proposed DNN method without sigma-based features was 79.04%.