|
|
Deep Learning for Medical Image Analysis
Trávníčková, Kateřina ; Hradiš, Michal (referee) ; Španěl, Michal (advisor)
This bachelor thesis deals with medical volume data analysis using convolutional neural networks. The input of the analysis is a CT scan of human limbs and the output are segmented countours of long bones, humerus and tibia. The goal of this work is to find suitable convolutional neural network settings to achieve the best possible analysis output while the area under the Precision-Recall curve is used as the precision metric. The best accuracy reaches almost 88 % (0.8778 AUC). The implementation is based on Caffe framework, or python caffe module.
|
|
|
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.
|
|
|
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.
|
|
|
Effect of practice on changing the work motivation and identification with the social role of the nurse
Trávníčková, Kateřina ; Štefančíková, Mariana (advisor) ; Vaňková, Dana (referee)
Name and surname of the author: Kateřina Trávníčková Institution: Charles University in Prague, Faculty of Medicine in Hradec Králové, Department of Social Medicine, Department of Nursing Title: Effect of practice on changing the work motivation and identification with the social role of the nurse Supervisor: PhDr. Mariana Štefačíková, Ph. D. Number of pages: 131 Year of defense: 2019 Keywords: Motivation, Work motivation, motivation factors, social role, identification with social role The bachelor thesis deals with motivation to work of nurse, motivation factors and social role of nurse. The theoretical part focuses on the concept of motivation, work motivation, the theory of work motivation and determinants affecting the nurse's motivation. It also focuses on the notion of social role and identification with the social role of a nurse. The research of motivation to work and the social role of the nurse is done by the method of semi-structured interviews, which are supplemented by a questionnaire survey. The obtained data are processed using open coding and triangulation (interviews, questionnaire survey, survey of the Ministry of Health of the Czech Republic). Here the nurses describe their original and current motivation to work as a nurse, factors that could jeopardize their retention, the...
|
|
|
Deep Learning for Medical Image Analysis
Trávníčková, Kateřina ; Hradiš, Michal (referee) ; Španěl, Michal (advisor)
This bachelor thesis deals with medical volume data analysis using convolutional neural networks. The input of the analysis is a CT scan of human limbs and the output are segmented countours of long bones, humerus and tibia. The goal of this work is to find suitable convolutional neural network settings to achieve the best possible analysis output while the area under the Precision-Recall curve is used as the precision metric. The best accuracy reaches almost 88 % (0.8778 AUC). The implementation is based on Caffe framework, or python caffe module.
|
|
| |
guest ::
RSS feed.