|
|
Application of graph theory to the design and optimization data network architectures
Římský, Adam ; Trzos, Michal (referee) ; Novotný, Vít (advisor)
This masters'sthesis deals with graph theory and utilization of this theory for design and optimization of data network structures. Introduction chapter describes graph theory in general view, i.e. fundamental terms used for graph description, graph distinguishing, etc. Next part describes graph algorithms, for example a shortest path finding. After this I write about actual routing protocols where the graph algorithms are used. Last but one part deals with queuing theory and final part describes practical presentation of using graph theory for design and optimization of data network structure in Matlab programme environment.
|
|
|
Segmentation and density estimation in femur
Horáček, Jan ; Maršálek, Lukáš (advisor)
Title: Segmentation and Density Estimation in Femur Author: Jan Horáček Department: Department of Software and Computer Science Education Supervisor: Mgr. Lukáš Maršálek Supervisor's e-mail address: lukas.marsalek@mff.cuni.cz Abstract: There are two basic ways of surgical treatment of femoral neck frac- ture - total endoprosthesis or a special metal screw in the bone. In case of a dense enough bone the screws are preferred. But for a too sparse bone, the screw may cut through the femoral head, which results in further surgical operations. We present a novel method for femoral head density measurement, which serves as another hint for the doctor's decision, whether to apply a screw or total endo- prosthesis. Our approach is based on semi-automatical femoral head segmenta- tion from CT dataset based on finding optimal path through polar coordinates on axial slices. The cost function is based on a combination of corticallis proper- ties, mostly the directional behavior of 3D gradients and their size in 2D slices, where they form typical "channels". The final volume is computed using filling and morphological algorithms and its properties are further measured. The final implementation was experimentally validated on RTG clinic of Bulovka hospital and allows radiologists to intuitively and accurately estimate...
|
|
|
Systém pro návrh optických sítí FTTH
Starý, Miroslav
The thesis explains the principle of optical conduction and describes its possible structures. It also deals with the automation of designing, projecting, constructing FTTH optical networks, processing RUIAN data, creating plugins into the QGIS geographic information system and implementing them using the Python programming language.
|
|
|
Segmentation and density estimation in femur
Horáček, Jan ; Maršálek, Lukáš (advisor)
Title: Segmentation and Density Estimation in Femur Author: Jan Horáček Department: Department of Software and Computer Science Education Supervisor: Mgr. Lukáš Maršálek Supervisor's e-mail address: lukas.marsalek@mff.cuni.cz Abstract: There are two basic ways of surgical treatment of femoral neck frac- ture - total endoprosthesis or a special metal screw in the bone. In case of a dense enough bone the screws are preferred. But for a too sparse bone, the screw may cut through the femoral head, which results in further surgical operations. We present a novel method for femoral head density measurement, which serves as another hint for the doctor's decision, whether to apply a screw or total endo- prosthesis. Our approach is based on semi-automatical femoral head segmenta- tion from CT dataset based on finding optimal path through polar coordinates on axial slices. The cost function is based on a combination of corticallis proper- ties, mostly the directional behavior of 3D gradients and their size in 2D slices, where they form typical "channels". The final volume is computed using filling and morphological algorithms and its properties are further measured. The final implementation was experimentally validated on RTG clinic of Bulovka hospital and allows radiologists to intuitively and accurately estimate...
|
|
|
Application of graph theory to the design and optimization data network architectures
Římský, Adam ; Trzos, Michal (referee) ; Novotný, Vít (advisor)
This masters'sthesis deals with graph theory and utilization of this theory for design and optimization of data network structures. Introduction chapter describes graph theory in general view, i.e. fundamental terms used for graph description, graph distinguishing, etc. Next part describes graph algorithms, for example a shortest path finding. After this I write about actual routing protocols where the graph algorithms are used. Last but one part deals with queuing theory and final part describes practical presentation of using graph theory for design and optimization of data network structure in Matlab programme environment.
|
|
| |
guest ::
