Skoči na vsebino
DomovPredmetiZajem in računalniško podprta analiza slik

Zajem in računalniško podprta analiza slik

Vrsta predmeta: teoretični

Število kreditnih točk: 5

Semester izvajanja: Celoletni

Opis predmeta

– Zajemanje digitalnih slik: optično-zaznavne lastnosti človeškega vida, vrste in definicije digitalnih slik, predstavitve barv in barvni prostori, parametri kakovosti, tehnologije zajemanja vizualnih podatkov na makro in mikroskopski ravni z digitalnimi fotoaparati in kamerami, osnove rentgenskega slikanja, računalniške tomografije, magnetne resonance in ultrazvoka, razumevanje vsebine slik.

– Prikazovanje, manipulacija in zgoščevanje sivinskih, barvnih in večdimenzionalnih slik

– Analiza slik: upragovanje, opisovanje s poravnavo topoloških, fizikalnih ali statističnih modelov, regresija in razvščanje slik z globokimi samoučečimi modeli, izločanje značilnic objektov zanimanja, analiza rasti in gibanja.

– Načrtovanje in uporaba slikovnih informacijskih sistemov: programska orodja za pridobivanje in analizo slik, načrtovanje, integracija in uporaba slikovnih informacijskih sistemov v biotehniških raziskavah in aplikacijah (mikroskopija, kontrola kakovosti živil, spremljanje rasti in gibanja živali, rastlin in mikroorganizmov, itn).

Cilji in kompetence

Seznaniti študente s področjem zajemanja in računalniško podprte analize biomedicinskih slik; posredovati znanje o sodobnih postopkih za zajemanje biomedicinskih slik, za njihovo prikazovanje, manipulacijo, zgoščevanje, ter kvantitativno analizo; posredovati znanje o strojnem in globokem strojnem učenju in uporabo teh orodij za regresijo in razvrščenje na podlagi biomedicinskih slik ter njihovo analizo; seznanjanje s pristopi k načrtovanju in uporabi slikovnih informacijskih sistemov v biotehniških raziskavah in aplikacijah.

Metode poučevanja in učenja

Teoretične osnove in širši pregled nad področjem predmeta študentje pridobijo na predavanjih, praktična znanja in izkušnje pa pri laboratorijskih vajah  in izdelavi izbrane projektne ali seminarske naloge z njihovega področja zanimanja.

Predvideni študijski rezultati

Študenti, ki bodo izbrali ta predmet, bodo pridobili znanja o zajemanju digitalnih slik; znali prikazovati, manipulirati in zgoščevati slike; znali izbrati in uporabljati obstoječe postopke; digitalne analize slik; znali načrtovati, učiti in vrednotiti globoke samoučeče modele za regresijo in razvrščanje na podlagi slikovne informacije; znali načrtovati in uporabljati slikovne informacijske sisteme v bioznanostih.

Reference nosilca

Žiga Špiclin

  1. BIZJAK, Žiga, PERNUŠ, Franjo, ŠPICLIN, Žiga. Deep shape features for predicting future intracranial aneurysm growth. Frontiers in physiology, vol. 12, article 644349, str. 1-10, 202
  2. SAVŠEK, Lina, STERGAR, Tamara, STROJNIK, Vojko, IHAN, Alojz, KOREN, Aleš, ŠPICLIN, Žiga, ŠEGA, Saša. Impact of aerobic exercise on clinical and magnetic resonance imaging biomarkers in persons with multiple sclerosis : an exploratory randomized controlled trial. Journal of rehabilitation medicine, vol. 53, iss. 4, str. 1-9, 2021.
  3. MADAN, Hennadii, BERLOT, Rok, RAY, Nicola J., PERNUŠ, Franjo, ŠPICLIN, Žiga. Practical priors for Bayesian inference of latent biomarkers. IEEE journal of biomedical and health informatics, vol. 24, no. 2, str. 396-406, 2020.
  4. JERMAN, Tim, CHIEN, Aichi, PERNUŠ, Franjo, LIKAR, Boštjan, ŠPICLIN, Žiga. Automated cutting plane positioning for intracranial aneurysm quantification. IEEE transactions on bio-medical engineering, vol. 67, no. 2, str. 577-587, 2020.
  5. MADAN, Hennadii, PERNUŠ, Franjo, ŠPICLIN, Žiga. Reference-free error estimation for multiple measurement methods. Statistical methods in medical research, vol. 28, issue 7, str. 2196-2209, 2019.
  6. MITROVIĆ, Uroš, LIKAR, Boštjan, PERNUŠ, Franjo, ŠPICLIN, Žiga. 3D-2D registration in endovascular image-guided surgery : evaluation of state-of-the-art methods on cerebral angiograms. International journal of computer assisted radiology and surgery : a journal for interdisciplinary research, developemnt and applications of image guided diagnosis and therapy, vol. 13, no. 2, str. 193-202, 2018.

 

Tomaž Vrtovec

  1. KOREZ, Robert, PUTZIER, Michael, VRTOVEC, Tomaž. A deep learning tool for fully automated measurements of sagittal spinopelvic balance from X-ray images : performance evaluation. European spine journal, vol. 29, no. 9, str. 2295-2305, 2020.
  2. KHOLIAVCHENKO, M., SIRAZITDINOV, I., KUBRAK, K., BADRUTDINOVA, R., KULEEV, R., YUAN, Y., VRTOVEC, Tomaž, IBRAGIMOV, Bulat. Contour-aware multi-label chest X-ray organ segmentation. International journal of computer assisted radiology and surgery : a journal for interdisciplinary research, developemnt and applications of image guided diagnosis and therapy, vol. 15, iss. 4, str. 425-436, 2020.
  3. BRINK, Rob C., VAVRUCH, Ludvig, SCHLÖSSER, Tom P. C., ABUL-KASIM, Kasim, OHLIN, Acke, TROPP, Hans, CASTELEIN, René M., VRTOVEC, Tomaž. Three-dimensional pelvic incidence is much higher in (thoraco)lumbar scoliosis than in controls. European spine journal, vol. 28, no. 3, str. 544-550, 2019.
  4. KNEZ, Dejan, NAHLE, Imad S., VRTOVEC, Tomaž, PARENT, Stefan, KADOURY, Samuel. Computer-assisted pedicle screw trajectory planning using CT-inferred bone density : a demonstration against surgical outcomes. Medical Physics, vol. 46, no. 8, str. 3543-3554, 2019.
  5. PINHEIRO, Alan Petrônio, COELHO, Júlio Cézar, PASCHOARELLI VEIGA, Antônio C., VRTOVEC, Tomaž. A computerized method for evaluating scoliotic deformities using elliptical pattern recognition in X-ray spine images. Computer methods and programs in biomedicine, vol. 161, str. 85-92, 2018.
  6. MOČNIK, Domen, IBRAGIMOV, Bulat, XING, Lei, STROJAN, Primož, LIKAR, Boštjan, PERNUŠ, Franjo, VRTOVEC, Tomaž. Segmentation of parotid glands from registered CT and MR images. Physica medica, vol. 52, str. 33-41, 2018.

 

Boštjan Likar

  1. ZELINSKYI, Yevhen, NAGLIČ, Peter, PERNUŠ, Franjo, LIKAR, Boštjan, BÜRMEN, Miran. Fast and accurate Monte Carlo simulations of subdiffusive spatially resolved reflectance for a realistic optical fiber probe tip model aided by a deep neural network. Biomedical optics express, vol. 11, no. 7, str. 3875-3889, 2020.
  2. MEHLE, Andraž, KITAK, Domen, PODREKAR, Gregor, LIKAR, Boštjan, TOMAŽEVIČ, Dejan. In-line agglomeration degree estimation in fluidized bed pellet coating processes using visual imaging. International journal of pharmaceutics, vol. 546, no. 1/2, str. 78-85, 2018.
  3. NAGLIČ, Peter, PERNUŠ, Franjo, LIKAR, Boštjan, BÜRMEN, Miran. Lookup table-based sampling of the phase function for Monte Carlo simulations of light propagation in turbid media. Biomedical optics express, vol. 8, no. 3, str. 1895-1910, 2017.
  4. IVANČIČ, Matic, NAGLIČ, Peter, PERNUŠ, Franjo, LIKAR, Boštjan, BÜRMEN, Miran. Virtually increased acceptance angle for efficient estimation of spatially resolved reflectance in the subdiffusive regime : a Monte Carlo study. Biomedical optics express, vol. 8, no. 11, str. 4872-4886, 2017.
  5. MEHLE, Andraž, LIKAR, Boštjan, TOMAŽEVIČ, Dejan. In-line recognition of agglomerated pharmaceutical pellets with density-based clustering and convolutional neural network. IPSJ transactions on computer vision and applications, vol. 9, 7, str. 1-6, 2017.
  6. JEMEC, Jurij, PERNUŠ, Franjo, LIKAR, Boštjan, BÜRMEN, Miran. Three-dimensional point spread function measurements of imaging spectrometers. Journal of optics, no. 9, 095002, str. 1-7, 2017.

Franjo Pernuš

  1. IVANČIČ, Matic, NAGLIČ, Peter, PERNUŠ, Franjo, LIKAR, Boštjan, BÜRMEN, Miran. Efficient estimation of subdiffusive optical parameters in real time from spatially resolved reflectance by artificial neural networks. Optics letters, vol. 43, no. 12, str. 2901-2904, 2018.
  2. LESJAK, Žiga, GALIMZIANOVA, Alfiia, KOREN, Aleš, LUKIN, Matej, PERNUŠ, Franjo, LIKAR, Boštjan, ŠPICLIN, Žiga. A novel public MR image dataset of multiple sclerosis patients with lesion segmentations based on multi-rater consensus. Neuroinformatics, vol. 16, no. 1, str. 51-63, 2018.
  3. IBRAGIMOV, Bulat, KOREZ, Robert, LIKAR, Boštjan, PERNUŠ, Franjo, XING, Lei, VRTOVEC, Tomaž. Segmentation of pathological structures by landmark-assisted deformable models. IEEE transactions on medical imaging, vol. 36, no. 7, str. 1457-1469, 2017.
  4. JEMEC, Jurij, PERNUŠ, Franjo, LIKAR, Boštjan, BÜRMEN, Miran. 2D sub-pixel point spread function measurement using a virtual point-like source. International journal of computer vision, vol. 121, no. 3, str. 391-402, 2017.
  5. MADAN, Hennadii, PERNUŠ, Franjo, LIKAR, Boštjan, ŠPICLIN, Žiga. A framework for automatic creation of gold-standard rigid 3D-2D registration datasets. International journal of computer assisted radiology and surgery : a journal for interdisciplinary research, developemnt and applications of image guided diagnosis and therapy, vol. 12, no. 2, str. 263-275, 2017.
  6. AKSOY, T., ŠPICLIN, Žiga, PERNUŠ, Franjo, UNAL, Gozde. Monoplane 3D-2D registration of cerebral angiograms based on multi-objective stratified optimization. Physics in Medicine & Biology, vol. 62, no. 24, str. 9377-9394, 2017.

Temeljni viri in literatura

– Thomas M. Deserno. Biomedical Image Processing. Springer, 2011.

– Klaus D. Tonnies. Guide to Medical Image Analysis: Methods and Algorithms. Springer, 2012.

– Deep Learning (Ian J. Goodfellow, Yoshua Bengio and Aaron Courville), MIT Press, 2016.

– Boštjan Likar. Biomedicinska slikovna informatika in diagnostika, 1. izdaja, Založba FE in FRI,

Ljubljana: Fakulteta za elektrotehniko, 2008.

Bodi na tekočem

Univerza v Ljubljani, Fakulteta za elektrotehniko, Tržaška cesta 25, 1000 Ljubljana

E:  dekanat@fe.uni-lj.si T:  01 4768 411

Vprašajte nas