Luka Šajn was born on 24.10.1977 in Ljubljana, Slovenia. He finished his PhD studies from the University of Ljubljana in 2007 with the thesis "Multi-resolution parametrization for texture classification and its application in analysis of scintigraphy images". His research interests are in: cell counting, multi-resolution pattern parametrization, ischaemic heart-disease diagnosing from scintigraphic images and whole-body bone scintigraphy segmentation. Currently he is a researcher and an assistant professor at the same faculty.
Activities and Events
Multi-resolution parametrization for texture classification and its application in analysis of scintigraphy images
The dissertation contributes to the new approaches in the major area of computer science, specifically in the domain field of machine learning, computer vision, texture parametrization and intelligent data analysis in medical applications.
In the dissertation multiresolutional texture parametrization is addressed and the original algorithm ARes for finding more informative resolutions in the sense of classification accuracy is proposed. ARes is designed to be used in combination with the existing parametrization algorithm ArTex developed by Bevk (2005). The results obtained using the ArTex parametrization algorithm in combination with ARes are compared with standard parametrization methods such as Gabor filters, Haar and Laws wavelets and Image Processor.
Our study explores the multiresolutional texture parametrization approach based on the image content with regard to the parametrization quality, especially in case of the ArTex algorithm. The tested parametrization algorithms (geometric algorithms, signal processing methods and statistical methods) using multiresolutional approach have demonstrated significant improvements in results over one scale parametrization. This supports the hypothesis that the resolution selection is important for texture parametrization. The developed algorithm ARes in combination with the ArTex algorithm has shown to be an appropriate tool as it achieves statistically significant improvements over single resolution and also over equidistant resolutions. The algorithm ARes in many cases also improves the performance of other parametrization algorithms in comparison to single resolution approach, whereas compared to the equidistant resolution approach it usually shows no significant improvement. We have confirmed that the use of the equidistant resolution space when parameterizing textures significantly outperforms the use of the exponential resolution space, which is used by majority of authors.
For the multiresolution parametrization applicative domain two medical cases have been used, sequential diagnostics of coronary artery disease and diagnostics of whole-body bone scintigraphy.
- Artificial intelligence and intelligent systems, Research Programme, P2-0209, 2004−2014
- Machine Learning of Probabilities with Applications to Web Portals and Medical Diagnostics, Bilateral Collaboration Project, BI-PT/06-07-004, 2006−2008
- Reliable and Comprehensible Machine Learning Approaches with Applications to Medical Diagnostics and Bioinformatics Bilateral, Bilateral Collaboration Project, 2005−2007
LOJK, Jasna, ČIBEJ, Uroš, KARLAŠ, David, ŠAJN, Luka, PAVLIN, Mojca. Comparison of two automatic cell-counting solutions for fluorescent microscopic images. Journal of Microscopy, ISSN 0022-2720, 2015, vol. , no. , str. 1-10,
PEER, Peter, JAKLIČ, Aleš, ŠAJN, Luka. A computer vision based system for a rehabilitation of a human hand. Periodicum biologorum, ISSN 0031-5362, 2013, vol. 115, no. 4, str. 535-544,
ŠAJN, Luka, KUKAR, Matjaž. Image processing and machine learning for fully automated probabilistic evaluation of medical images. Computer methods and programs in biomedicine, ISSN 0169-2607. [Print ed.], Dec. 2011, vol. 104, no. 3, str. 75-86,
ŠAJN, Luka, KUKAR, Matjaž. Image processing and machine learning for fully automated probabilistic evaluation of medical images. Comput. methods programs biomed.. [Print ed.], str. [1-12]
ŠAJN, Luka, KONONENKO, Igor. Image segmentation and parameterization for automatic diagnostics of whole-body scintigrams : basic concepts. V: SCHAEFER, Gerald (ur.), HASSANIEN, Aboul Ella (ur.), JIANG, J. (ur.). Computational intelligence in medical imaging : techniques and applications. Boca Raton; London; New York: CRC Press; Taylor & Francis Group, cop. 2009, str. 347-377,  str.
ŠAJN, Luka, KONONENKO, Igor. Multiresolution image parametrization for improving texture classification. EURASIP J. Adv. Signal Process. (Print). [Print ed.], 2008, vol. 2008, no. , str. 1-12,
ŠAJN, Luka, KONONENKO, Igor, MILČINSKI, Metka. Computerized segmentation and diagnostics of whole-body bone scintigrams. Comput. med. imaging graph.. [Print ed.], 2007, vol. 31, no. 7, str. 531-541,
ŠAJN, Luka, KUKAR, Matjaž, KONONENKO, Igor, MILIČINSKI, Metka. Computerized segmentation of whole-body bone scintigrams and its use in automated diagnostics. Comput. methods programs biomed.. [Print ed.], 2005, vol. 80, no. 1, str. -55,
KUKAR, Matjaž, ŠAJN, Luka, GROŠELJ, Ciril, GROŠELJ, Jera. Multi-resolution image parametrization in stepwise diagnostics of coronary artery disease. Lect. notes comput. sci., str. -129,
ŠAJN, Luka, KUKAR, Matjaž, KONONENKO, Igor, MILIČINSKI, Metka. Automatic segmentation of whole-body bone scintigrams as a preprocessing step for computer assisted diagnostics. Lect. notes comput. sci., str. -372,
ŠAJN, Luka, BEVK, Matjaž, KONONENKO, Igor, MILIČINSKI, Metka. Automatic diagnostics of whole-body scintigrams using image segmentation and parametrization. Eur. j. nucl. med. mol. imaging (Print), Sep. 2005, vol. 32, suppl. 1, str. 265. [COBISS.SI-ID4985940]