When using pre-trained models, only the last layer or two of the CNN are re-trained on the dataset of interest, but the rest of the model is kept unchanged. Many of these models are pre-trained on the dataset from the ImageNet Large Scale Visual Recognition Challenge (ILSVRC) containing millions of non-medical images. There are currently many readily available open-source implementations of CNNs through frameworks such as Caffe. Īn alternative to training a CNN de novo is to use pre-trained CNN models. By using a large dataset of ~ 53,000 studies and training the novel DenseNet CNN de novo, one study achieved a 97% fracture detection accuracy for the femoral neck on pelvic radiographs. Rather than requiring manual feature engineering of the images, convolution neural networks (CNNs) allow for image evaluation with native image inputs. One study performed manual feature extraction on 145 radiographs, utilized a random forest machine learning algorithm, and achieved a fracture detection accuracy of 81%. Several recent studies have demonstrated the utility of machine learning for fracture detection in musculoskeletal images. Despite our small dataset size, by utilizing an ensemble of models and 3 views for each case, we achieved an accuracy of 81%, which was in line with the accuracy of other models using a much higher number of cases with pre-trained models and models which implemented manual feature extraction. When all three views for a single case were utilized, the ensemble of all models resulted in the best output metrics with an accuracy of 81%. For single radiographic views, the ensemble of all 5 models produced the best accuracy at 76%. A voting method was implemented to consolidate the output from the three views and model ensemble. Ensembles were created from a combination of CNNs after training. Model outputs were evaluated using both one and three radiographic views. Measured output metrics were accuracy, positive predictive value (PPV), negative predictive value (NPV), sensitivity, and specificity. Training was performed using single radiographic views. The Inception V3, Resnet, and Xception convolutional neural networks were constructed utilizing the Python programming language with Tensorflow as the framework. Data augmentation was performed during training. Single- and multiview models were created to determine the effect of multiple views. To determine whether we could train convolutional neural network (CNN) models de novo with a small dataset, a total of 596 normal and abnormal ankle cases were collected and processed.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |