DEDAD - Strukturne dekompozicije empirijskih podataka za računalno potpomognutu dijagnostiku bolesti
Kategorija
Projekti Hrvatske zaklade za znanost
Iznos financiranja
594.400,00
Datum početka
1.3.2017.
Datum završetka
27.2.2021.
Status
Završen
Glavni istraživač
I
K
I
K
Ciljevi projekta su razvoja algoritama za strukturne dekompozicije empirijskih podataka za računalno potpomognutu dijagnostiku bolesti koristeći:
- sliku nebojenih i sliku preparat bojenih hameleon-eosinom (H&E) prikupljenih analizom tkiva na smrznutim rezovima, odnosno uzorcima humane jetre s metastazama tumora debelog crijeva.
- metaboličko profiliranje u studiji identifikacije biomarkera iz 1H NMR spektara uzoraka humanog urina pozitivnih i negativnih na dijabetes II.
- sliku optičke koherentne tomografije retine (mrežnice).
Za ostvarenje prethodno navedenih ciljeva na projektu će se raditi na razvoju algoritama za
- nenadzirano grupiranje podataka u (nisko dimenzionalnim) podprostorima sa naglaskom na strukturno ograničeno učenje reprezentacije empirijskih skupova podataka.
- polu-nadzirane multiplikativne faktorizacije nenegativnih matrica u Hilbertovim prostorima induciranim višestrukim jezgrama.
- fuziju RGB slike sa ciljem poboljšanja i standardizacije kvalitete.
- aditivne strukturno ograničene faktorizacije nenegativnih matrica.
Publikacije
Radovi u znanstvenim časopisima
- D. Sitnik, G. Aralica, M. Hadžija, M. Popović Hadžija, A. Pačić, M. Milković Periša, L. Manojlović, K. Krstanac, A. Plavetić, I. Kopriva (2021), "A Dataset and a Methodology for Intraoperative Computer-Aided Diagnosis of a Metastatic Colon Cancer in a Liver," Biomedical Signal Processing and Control, vol. 66, April 2021, article no. 102402. IF: 3.137, Q2: leading 37% (32/87) in Biomedical Engineering. https://doi.org/10.1016/j.bspc.2020.102402 Dataset: http://cocahis.irb.hr
- I. Kopriva, I. Jerić, M. P. Hadžija, M. Hadžija, M. Vučić Lovrenčić (2021), "Nonnegative Least Squares Approach to Quantification of 1H Nuclear Magnetic Resonance Spectra of Human Urine," Analytical Chemistry, vol. 93, no. 2, pp. 745-751, 2021. IF: 6.785 - Q1: leading 8% (7/86) in Analytical Chemistry. https://doi.org/10.1021/acs.analchem.0c02837
- M. Brbić, I. Kopriva (2020), "l0 Motivated Low-Rank Sparse Subspace Clustering," IEEE Transactions on Cybernetics, vol. 50, no. 4, pp. 1711-1725, https://doi.org/10.1109/TCYB.2018.2883566 (IF: 11.079) - Q1: leading 4% (5/136) in Computer Science, Artificial Intelligence. Matlab code: https://github.com/mbrbic/L0-Motivated-LRSSC
- I. Kopriva, I. Jerić, M. Popović Hadžija, M. Hadžija. M. Vučić Lovrenčić, L. Brkljačić (2019), "Library-Assisted Nonlinear Underdetermined Blind Separation and Annotation of Pure Components from 1H Nuclear Magnetic Resonance Mixture Spectra," Analytica Chimica Acta, vol. 1080, pp. 55-65, 2019. https://doi.org/10.1016/j.aca.2019.07.004 (IF: 5.256) - Q1: leading 12% (10/84) in Analytical Chemistry.
- D. Tolić, N. Antulov Fantulin, I. Kopriva (2018), " Non-negative Subspace Clustering in Nonlinear Orthogonal Non-negative Matrix Factorization Framework," Pattern Recognition, vol. 82, October 2018, pp. 40-55, https://doi.org/10.1016/j.patcog.2018.04.029 (IF: 5.898) - Q1: leading 11% (14/133) in Computer Science, Artificial Intelligence. Matlab code: https://github.com/singularity4/NonlinearOrthogonalNMF
- M. Brbić, I. Kopriva (2018), "Multi-view Low-rank Sparse Subspace Clustering," Pattern Recognition, vol. 73, January 2018, pp. 247-258, https://doi.org/10.1016/j.patcog.2017.08.024, (IF: 5.898) - Q1: leading 11% (14/133) in Computer Science, Artificial Intelligence. Matlab code: https://github.com/mbrbic/MultiViewLRSSC
- I. Kopriva, W. Ju, B. Zhang, F. Shi, D. Xiang, K. Yu, X. Wang, U. Bagci and X. Chen (2017), "Single-channel Sparse Nonnegative Blind Source Separation Method for Automatic 3D Delineation of Lung Tumor in PET Images," IEEE Journal of Biomedical and Health Informatics, vol. 21, No. 6, pp. 1656-1666, https://doi.org/10.1109/JBHI.2016.2624798, (IF: 3.85) - Q1: leading 12% (18/146) in Computer Science, Information Systems.
Radovi na znanstvenim skupovima:
- I. Kopriva, D. Sitnik, G. Aralica, A. Pačić, M. Popović Hadžija, M. Hadžija (2021), "Approximate explicit feature map for computational augmentation of quasi hyperspectral images from RGB images of hematoxylin and eosin stained histopathological specimens," Digital Pathology Conference - SPIE Medical Imaging Symposium 2021, Proc. 11603, article no. 116030R, https://doi.org/10.1117/12.2579408, 15.- 19., 2021, San Diego, USA.
- L. Tian, Q. Du, I. Kopriva (2020), "L0-Motivated Low-Rank Sparse Subspace Clustering for Hyperspectral Imagery (2020)," IGARSS 2020 - 2020 IEEE International Geoscience and Remote Sensing Symposium, pp. 1038-1041, https://doi.org/10.1109/IGARSS39084.2020.9324155, 26. 9. - 2. 10, 2020, Waikoloa HI, USA.
- I. Kopriva, F. Shi, M. Štanfel, X. Chen (2020), "Enhanced low-rank plus group sparse decomposition for speckle reduction in OCT images," Image Processing Conference - SPIE Medical Imaging Symposium 2020, Proc. 11313, article no. 113132H, https://doi.org/10.1117/12.2538466, 15.-20. February, 2020, Houston, USA.
- D. Sitnik, I. Kopriva, G. Aralica, A. Pačić, M. Popović Hadžija, M. Hadžija (2020), "Deep learning approaches for intraoperative pixel-based diagnosis of colon cancer metastasis in a liver from phase-contrast images of unstained specimens," Digital Pathology Conference - SPIE Medical Imaging Symposium 2020, Proc. 11320, article no. 1132009, https://doi.org/10.1117/12.2542799, 15.-20. February, 2020, Houston, USA.
- D. Sitnik, I. Kopriva, G. Aralica, A. Pačić, M. Popović Hadžija, M. Hadžija (2020), "Transfer Learning Approach for Intraoperative Pixel-based Diagnosis of Colon Cancer Metastasis in a Liver from Hematohylin-Eosin Stained Specimens," Digital Pathology Conference - SPIE Medical Imaging Symposium 2020, Proc. 11320, article no. 113200A, https://doi.org/10.1117/12.2538303, 15.-20. February, 2020, Houston, USA.
- L. Tian, Q. Du, I. Kopriva, and N. Younan (2019), "Orthogonal Graph-regularized Nonnegative Matrix Factorization for Hyperspectral Image Clustering," IGARSS 2019 - 2019 IEEE Geoscience and Remote Sensing Symposium, pp. 795-798, 28. 7. - 2. 8, 2019, Yokohama, Japan,https://doi.org/10.1109/IGARSS.2019.8897876
- I. Kopriva, G. Aralica, M. Popović Hadžija, M. Hadžija, L. I. Dion-Bertrand, X. Chen (2019), "Hyperspectral imaging for intraoperative diagnosis of colon cancer metastasis in a liver," SPIE Medical Imaging 2019 - Digital Pathology Conference, Proc. 10956, article no. 109560S, https://doi.org/10.1117/12.2503907, editors J. A. Tomaszewski, A. D. Ward, February 16 - 21, 2019, San Diego, CA, USA.
- I. Kopriva (2018), "Joint Nonnegative Matrix Factorization for Underdetermined Blind Source Separation in Nonlinear Mixtures,"14th International Conference on Latent Variable Analysis and Signal Separation(LVA ICA 2018), July 2-6, 2018, Guilford, UK., Springer International Publishing AG, part of Springer Nature 2018 Y. Deville. S. Gannot, R. Mason, M. D. Plumbly, D. Ward (Eds.): LVA/ICA 2018, LNCS 10891, pp. 107–115, 2018. https://doi.org/10.1007/978-3-319-93764-9_11
- I. Kopriva, M. Brbić, D. Tolić, N. Antulov-Fantulin, X. Chen (2017), "Fast Clustering in Linear 1D Subspaces: Segmentation of Microscopic Image of Unstained Specimen," SPIE Medical Imaging Symposium 2017 - Digital Pathology Conference, vol. 10140, http://dx.doi.org/10.1117/12.2247806, editor Metin N. Gurcan, John E. Tomaszewski, Orlando, US, February 11 - 16, 2017.