|
|
Deep Learning for Medical Image Analysis
Dronzeková, Michaela ; Kodym, Oldřich (referee) ; Španěl, Michal (advisor)
The purpose of this thesis is to use convolutional neural networks for X-ray image classification of human body. Four different architectures of neural networks have been created. They were trained and tested on three tasks: classification of front and lateral chest, classification of X-ray images into several different categories and classification of diseases in chest X-ray. ResNet and SEResNet architectures achieved the best results. SEResNet scored 99,49% accuracy in the first task, ResNet achieved 94,97% accuracy in the second task and SEResNet reached 31,53% in the third task with F1 measure as metrics for evaluating results.
|
|
|
Analysis of Polygonal Models Using Neural Networks
Dronzeková, Michaela ; Zemčík, Pavel (referee) ; Kodym, Oldřich (advisor)
This thesis deals with rotation estimation of 3D model of human jaw. It describes and compares methods for direct analysis od 3D models as well as method to analyze model using rasterization. To evaluate perfomance of proposed method, a metric that computes number of cases when prediction was less than 30° from ground truth is used. Proposed method that uses rasterization, takes three x-ray views of model as an input and processes it with convolutional network. It achieves best preformance, 99% with described metric. Method to directly analyze polygonal model as a sequence uses attention mechanism to do so and was inspired by transformer architecture. A special pooling function was proposed for this network that decreases memory requirements of the network. This method achieves 88%, but does not use rasterization and can process polygonal model directly. It is not as good as rasterization method with x-ray display, byt it is better than rasterization method with model not rendered as x-ray. The last method uses graph representation of mesh. Graph network had problems with overfitting, that is why it did not get good results and I think this method is not very suitable for analyzing plygonal model.
|
|
|
Analysis of Polygonal Models Using Neural Networks
Dronzeková, Michaela ; Zemčík, Pavel (referee) ; Kodym, Oldřich (advisor)
This thesis deals with rotation estimation of 3D model of human jaw. It describes and compares methods for direct analysis od 3D models as well as method to analyze model using rasterization. To evaluate perfomance of proposed method, a metric that computes number of cases when prediction was less than 30° from ground truth is used. Proposed method that uses rasterization, takes three x-ray views of model as an input and processes it with convolutional network. It achieves best preformance, 99% with described metric. Method to directly analyze polygonal model as a sequence uses attention mechanism to do so and was inspired by transformer architecture. A special pooling function was proposed for this network that decreases memory requirements of the network. This method achieves 88%, but does not use rasterization and can process polygonal model directly. It is not as good as rasterization method with x-ray display, byt it is better than rasterization method with model not rendered as x-ray. The last method uses graph representation of mesh. Graph network had problems with overfitting, that is why it did not get good results and I think this method is not very suitable for analyzing plygonal model.
|
|
|
Deep Learning for Medical Image Analysis
Dronzeková, Michaela ; Kodym, Oldřich (referee) ; Španěl, Michal (advisor)
The purpose of this thesis is to use convolutional neural networks for X-ray image classification of human body. Four different architectures of neural networks have been created. They were trained and tested on three tasks: classification of front and lateral chest, classification of X-ray images into several different categories and classification of diseases in chest X-ray. ResNet and SEResNet architectures achieved the best results. SEResNet scored 99,49% accuracy in the first task, ResNet achieved 94,97% accuracy in the second task and SEResNet reached 31,53% in the third task with F1 measure as metrics for evaluating results.
|
guest ::
