Research interests
I spend most of my time working on weakly-supervised learning from image datasets. I often employ a deep network in some way.
Contact
My email address is barucden [at] fel [•] cvut [•] cz.
My ORCID iD is 0000-0003-0428-3354.
Education
- 2019–present
- CTU in Prague, Ph.D., Machine learning with biomedical applications.
- 2017–2019
- CTU in Prague, Master (Ing.), Data science (with honors).
- 2014–2017
- CTU in Prague, Bachelor (Bc.), Software engineering (with honors).
Teaching experience
- Winter, 2022
- BAM33ZMO: Medical image analysis, 9 seminars.
- Winter, 2021
- B3B33ALP: Introduction to programming, 13 seminars.
- Winter, 2019 and 2020
- B4B33ALG: Basic data structures and algorithms, 13 seminars.
Publications
- D. Baručić, J. Kybic.
Learning to segment from object thickness annotations
. Accepted to ISBI 2023. - D. Baručić, et al.
Characterization of drug effects on cell cultures from phase-contrast microscopy images
. Computers in Biology and Medicine, 2022. - D. Baručić, J. Kybic.
Fast learning from label proportions with small bags
. International Conference on Image Processing (ICIP). IEEE, 2022. - D. Baručić, J. Kybic.
Learning to segment from object sizes
. Information Technologies – Applications and Theory (ITAT). CEUR-WS.org, 2022. - D. Baručić, et al.
Automatic evaluation of human oocyte developmental potential from microscopy images
. International Symposium on Medical Information Processing and Analysis (SIPAIM). SPIE, 2021.
Minor contributions
- S. Kaushik, et al.
The effect of hyperaldosteronism on carotid artery texture in ultrasound images
. Diagnostics. MDPI, 2022.