Thesis supervisor: András Hajdu
Location of studies (in Hungarian): University of Debrecen Faculty of Informatics Abbreviation of location of studies: DE IK
Description of the research topic:
Syllabus
The aim of the research is to develop methods for supporting clinical decisions and automatic screening systems primarily based on the processing of image/video data. Particularly, among the main tasks we can find the extraction of anatomic parts and lesions form images and videos recorded by mobile devices or high-tech clinical devices. Decision making can be realized using both traditional and advanced (e.g. deep learning) machine learning tools. As specific clinical fields, the research can focus on the processing of retinal or dermatologic images/videos, however, the development of general methodology is preferred. Efficient implementation issues are also to be investigated with a specific focus on Big Data related issues in case of large amount of visual data.
Bibliography
1. Yoshua Bengio, Ian J. Goodfellow, Aaron Courville: Deep Learning, MIT Press, 2015.
2. B. Cyganek: Object Detection and Recognition in Digital Images: Theory and Practice, John Wiley & Sons Ltd., New York, 2013.
3. R.O. Duda, P.E. Hart, D.G. Stork: Pattern Classification, John Wiley and Sons, 2001.
4. Handbook of Medical Imaging (I. Bankman ed.), Academic Press, San Diego, 2000.
Login
