|
|
Simulation of synthetic diffusion tensor data
Labudová, Kristýna ; Mézl, Martin (referee) ; Labounek, René (advisor)
This work deals with different approaches to imaging of diffusion intensity with magnetic resonance. Individual approaches are described and compared. Gaussian model for approximation of diffusion profile is analysed and mathematically determined in details. The next part of this work concerns about process of simulation synthetic diffusion tensor data, adding noise to data and estimation of diffusion tensor from noisy data. Estimation’s accuracy is rated according to deviation of fractional anisotropy of estimated and original tensor and also according to deviation of the main eigenvectors of both tensors. Accuracy of the estimation is evaluated automatically with the programme. There is realization of graphical interface for simulation as well as for automatical evaluation of results described in details. At the end of this work all results are processed and commented and there is also recommendation for optimal adjustment of the data acquisition. 120 gradient directions are the most optimal of all analysed direction. It provides sufficient accuracy of results with optimal time of data acquisition which is suitable for clinical praxis.
|
|
|
Image recognition for robotic hand
Labudová, Kristýna ; Jakubíček, Roman (referee) ; Harabiš, Vratislav (advisor)
This thesis concerns with processing of embedded terminals’ images and their classification. There is problematics of moire noise reduction thought filtration in frequency domain and the image normalization for further processing analyzed. Keypoints detectors and descriptors are used for image classification. Detectors FAST and Harris corner detector and descriptors SURF, BRIEF and BRISK are emphasized as well as their evaluation in terms of potential contribution to this work.
|
|
|
Quality characteristics of farmed and wildlife animals meat
LABUDOVÁ, Kateřina
Venison is currently considered a modern food, both because of its nutritional value, but also because of the ecological method of sourcing. In modern venison history, it has largely replaced livestock meat. This trend was also reflected in attempts to breed wild animals on farms. Farm breeding is also enjoying great success in the Czech Republic. Non-domesticated animals on farms are considered farm animals and their meat is referred to as farm animal meat - deer, fallow deer, mouflon etc. Farmed animals thus represent the possibility of year-round production of this meat as opposed to venison that, which is produced (hunted) for a limited period of time time of year. The aim of the work was to determine the differences in sensory properties and chemical composition between the meat (musculus longissimus dorsi) of males from the farmed breedeng of fallow deer (Dama dama), red deer (Cervus elaphus) and mouflon (Ovis musimon) slaughtered by shooting on the farm and meat of the same species originating from the wild of the Czech Republic. The survey used meat from individuals of the same age and condition with approximately the same time of slaughter at the turn of 2022/2023. The meat was tested by the evaluators using a sensory profile method. Fallow deer meat from wild was found to be a significantly higher (p < 0,05) intensity of color and intensity of veni-son flavour compared to fallow deer meat from farm, while red deer meat from farm was characterized by significantly higher (p < 0,05) pleasantness and color intensity and a higher intensity of venison flavour and also a higher juiciness, in contrast to wild red deer meat. The mouflon meat derived from a farm was found to be significantly higher (p < 0,05) intensity of color compared to wild mouflon meat. A higher palatability of meat were noted for red deer and fallow deer meat from farm, while mouflon meat from wild was marginally better assessed. Also found were differences in chemical composition animals from different production systems. Fallow deer meat from a farm contained significantly more (p < 0,05) of the total protein, while a higher intramuscular fat content was analyzed in farmed fallow deer meat. Red deer meat from a farm contained a significantly higher (p < 0,05) intramuscular fat and total protein content, while red deer venison analyzed had a significantly higher (p < 0,05) water content. A significantly higher (p < 0,05) total protein and collagen proteins content was found in wild mouflon meat.
|
|
|
Dopad různé alimentární expozice bisfenolu na játra potkanů
LABUDOVÁ, Kateřina
Bisphenol S (BPS) is an organic compound widely used in industrial production as a substitute for bisphenol A (BPA) for which negative impacts on the human body have been demonstrated and have therefore been significantly regulated. However, as time progresses, the negative properties of BPS are also shown, especially for the endocrine and reproductive systems. The liver is an organ in which xenobiotics are detoxified and thus BPS and BPA. The aim of the bachelor thesis was to verify whether 10-week oral exposure to BPS (4 g, 50 g, 100 mg/kg bw/day) and BPA (100 mg/kg bw/day) will lead to a negative impact on body weight, weight gain, weight and morphological liver structure and on the concentration of triglycerides, cholesterol, total white-vinas, urea and alkaline phosphatase activity in blood serum. The experiment involved 24 rats. BPA and BPS were administered daily by gastric tube in sunflower oil. The results of our research clearly did not show a negative effect of BPS and BPA in rats on body weight, absolute gain, absolute and relative liver weight in rats. There was also no evidence of histopathological changes in rat liver. In contrast, bio-chemical parameters were altered compared to the control group. Decreases in trigly-cerides, total protein content, urea, alkaline phosphatase activity and increases in serum cholesterol were noted. In the 50 g/kg body weight/day group of rats, the highest and lowest decreases in biochemical parameters were observed, respectively. The exposure dose of 100 mg BPS and BPA in our experiment did not cause liver histopathological changes or changes in biochemical parameters according to the hy-pothesis. A possible explanation could be given by the carrier for which sunflower oil with natural antioxidant content was used.
|
|
|
Image recognition for robotic hand
Labudová, Kristýna ; Jakubíček, Roman (referee) ; Harabiš, Vratislav (advisor)
This thesis concerns with processing of embedded terminals’ images and their classification. There is problematics of moire noise reduction thought filtration in frequency domain and the image normalization for further processing analyzed. Keypoints detectors and descriptors are used for image classification. Detectors FAST and Harris corner detector and descriptors SURF, BRIEF and BRISK are emphasized as well as their evaluation in terms of potential contribution to this work.
|
|
|
Simulation of synthetic diffusion tensor data
Labudová, Kristýna ; Mézl, Martin (referee) ; Labounek, René (advisor)
This work deals with different approaches to imaging of diffusion intensity with magnetic resonance. Individual approaches are described and compared. Gaussian model for approximation of diffusion profile is analysed and mathematically determined in details. The next part of this work concerns about process of simulation synthetic diffusion tensor data, adding noise to data and estimation of diffusion tensor from noisy data. Estimation’s accuracy is rated according to deviation of fractional anisotropy of estimated and original tensor and also according to deviation of the main eigenvectors of both tensors. Accuracy of the estimation is evaluated automatically with the programme. There is realization of graphical interface for simulation as well as for automatical evaluation of results described in details. At the end of this work all results are processed and commented and there is also recommendation for optimal adjustment of the data acquisition. 120 gradient directions are the most optimal of all analysed direction. It provides sufficient accuracy of results with optimal time of data acquisition which is suitable for clinical praxis.
|
|
|
Simulation of synthetic diffusion tensor data
Labudová, K.
First there is the Gaussian diffusion tensor model with the most important equations introduced. The other part of this work deals with the software and its current state of development. The software calculates diffusion tensor data from diffusion tensor and acquisition parameters successfully. User can add noise to simulated data and then estimate diffusion tensor in reverse. Original and estimated diffusion tensors are displayed, compared and their deviation in space determined.
|
guest ::
RSS feed.