|
|
Volumetric Segmentation of Dental CT Data
Berezný, Matej ; Kodym, Oldřich (referee) ; Čadík, Martin (advisor)
The main goal of this work was to use neural networks for volumetric segmentation of dental CBCT data. As a byproducts, both new dataset including sparse and dense annotations and automatic preprocessing pipeline were produced. Additionally, the possibility of applying transfer learning and multi-phase training in order to improve segmentation results was tested. From the various tests that were carried out, conclusion can be drawn that both multi-phase training and transfer learning showed substantial improvement in dice score for both sparse and dense annotations compared to the baseline method.
|
|
|
Interactive 3D CT Data Segmentation Based on Deep Learning
Trávníčková, Kateřina ; Hradiš, Michal (referee) ; Kodym, Oldřich (advisor)
This thesis deals with CT data segmentation using convolutional neural nets and describes the problem of training with limited training sets. User interaction is suggested as means of improving segmentation quality for the models trained on small training sets and the possibility of using transfer learning is also considered. All of the chosen methods help improve the segmentation quality in comparison with the baseline method, which is the use of automatic data specific segmentation model. The segmentation has improved by tens of percents in Dice score when trained with very small datasets. These methods can be used, for example, to simplify the creation of a new segmentation dataset.
|
|
|
Displaying 3D Graphics in Web Browser
Sychra, Tomáš ; Pečiva, Jan (referee) ; Španěl, Michal (advisor)
This thesis discusses possibilities of accelerated 3D scene displaying in a Web browser. In more detail, it deals with WebGL standard and its use in real applications. An application for visualization of volumetric medical data based on JavaScript, WebGL and Three.js library was designed and implemented. Image data are loaded from Google Drive cloud storage. An important part of the application is 3D visualization of the volumetric data based on volume rendering technique called Ray-casting.
|
|
|
Deep Learning for Medical Image Analysis
Osvald, Martin ; Juránek, Roman (referee) ; Španěl, Michal (advisor)
The goal of this bachelor's thesis is to use the 2D convolutional neural network on the 3D model dataset by multi-view methods. The view is 2D picture of 3D model. The result are Pyqt applications, where is possible to load the 3D model of teeth and predict the location of landmarks and teeth by object segmentation and object detection. During this thesis, an annotation's script was created for the annotation of 3D models for landmarks of teeth and teeth themself. This thesis solves the problem of the small availability of annotated 3D datasets in the medical industry by automating generating binary masks from different views on 3D models.
|
|
|
Rendering Volumetric Data in Web Browser
Fisla, Jakub ; Zemčík, Pavel (referee) ; Španěl, Michal (advisor)
This thesis discusses rendering capabilities of web browsers of accelerated 3D scene rendering. It specifically deals with direct volumetric medical data visualization. It focuses on the usage of ray casting algorithm, its quality and its realistic rendering options. One of the goals was to create an application that demonstrates the ability to render three-dimensional volume data in a web browser using WebGL. The application is written in JavaSript and its 3D rendering core uses the Three.js library.
|
|
|
Patient data trasfer over GSM
Pavliš, Jaroslav ; Dlouhý, Jiří (referee) ; Švrček, Martin (advisor)
This diploma thesis is concerned with possibilities of patient data transfer from a pacemaker or implantable cardioverter-defibrillator to physician over GSM. Theoretical part describes options of data transfer in GSM networks, data appropriate for sending and a structure of message is proposed. A device, that is able to send medical data in a form of SMS messages is designed and constructed. The device uses a Freescale MC68HC908GP32 microcontroller, character display with a Hitachi HD44780 controller and a cell phone Sony CMD-J70. The program for microcontroller is written in assembler for HC08. For tabular view of received messages, an application software for PC was created.
|
|
|
Displaying 3D Graphics in Web Browser
Sychra, Tomáš ; Kršek, Přemysl (referee) ; Španěl, Michal (advisor)
This bachelor's thesis deals with display of accelerated 3D graphics in a web browser environment. Existing technologies such as WebGL are presented and discussed. Further, in the second part of the thesis, an application for browsing medical volumetric data is designed and implemented. The application is built with the WebGL technology and Javascript graphics engine called O3D API.
|
|
|
Deep Learning for Medical Image Analysis
Osvald, Martin ; Juránek, Roman (referee) ; Španěl, Michal (advisor)
The goal of this bachelor's thesis is to use the 2D convolutional neural network on the 3D model dataset by multi-view methods. The view is 2D picture of 3D model. The result are Pyqt applications, where is possible to load the 3D model of teeth and predict the location of landmarks and teeth by object segmentation and object detection. During this thesis, an annotation's script was created for the annotation of 3D models for landmarks of teeth and teeth themself. This thesis solves the problem of the small availability of annotated 3D datasets in the medical industry by automating generating binary masks from different views on 3D models.
|
|
|
Volumetric Segmentation of Dental CT Data
Berezný, Matej ; Kodym, Oldřich (referee) ; Čadík, Martin (advisor)
The main goal of this work was to use neural networks for volumetric segmentation of dental CBCT data. As a byproducts, both new dataset including sparse and dense annotations and automatic preprocessing pipeline were produced. Additionally, the possibility of applying transfer learning and multi-phase training in order to improve segmentation results was tested. From the various tests that were carried out, conclusion can be drawn that both multi-phase training and transfer learning showed substantial improvement in dice score for both sparse and dense annotations compared to the baseline method.
|
|
|
Interactive 3D CT Data Segmentation Based on Deep Learning
Trávníčková, Kateřina ; Hradiš, Michal (referee) ; Kodym, Oldřich (advisor)
This thesis deals with CT data segmentation using convolutional neural nets and describes the problem of training with limited training sets. User interaction is suggested as means of improving segmentation quality for the models trained on small training sets and the possibility of using transfer learning is also considered. All of the chosen methods help improve the segmentation quality in comparison with the baseline method, which is the use of automatic data specific segmentation model. The segmentation has improved by tens of percents in Dice score when trained with very small datasets. These methods can be used, for example, to simplify the creation of a new segmentation dataset.
|
guest ::
