Detekcija jezgri stanica uz pomoć algoritama evolucijskog računanja

undergraduate thesis

• Title (english): Detection of cell nuclei using evolutionary computation
• Author: Josip Matak
• Mentor: Domagoj Jakobović (mentor)
• Committee member: Domagoj Jakobović (predsjednik povjerenstva)
• Committee member: Nikolina Frid (član povjerenstva)
• Committee member: Marko Đurasević (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2018-07-13, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Tema ovog rada je problem detekcije jezgara stanica iz mikroskopskih slika pristupom temeljenim na metodama morfoloških transformacija optimiziranih evolucijskim algoritmima i metodom odsijecanja praga. Problem se sveo na optimiranje pravilnog slijeda morfoloških transformacija i njima pripadnog strukturnog elementa korištenog u obradi slike. Ostvarena je mogućnost evoluiranja tog slijeda koristeći genetski algoritam i genetsko programiranje, čije su značajke opisane i uspoređivane. Dobiveni rezultati pokazali su kako model nije dovoljno prilagodljiv da bi nadmašio do sada korištene tehnike segmentacije biomedicinskih slika, premda se pokazao kao rješenje koje na okruženju koje nije dodatno zakomplicirano daje brze i točne rezultate. Takvi konačni rezultati jednim dijelom učinak su spore obrade velike količine slika i vremenske ograničenosti na dodatna istraživanja.

Abstract (english)

Main problem elaborated in this paper is detecting cell's nucleus within given microscopic photo using methods of morphology transformations optimized by evolutionary computing algorithms and thresholding methods for segmentation featuring elements. Optimization problem is finding the most suitable configuration of given morphology transformations and their given structuring element used in image segmentation. Realized versions of optimization techniques are genetic algorithm and genetic programming which features are compared in paper. Final results suggested that resulted model wasn't as successful as modern day used segmentation techniques. Although, in much simpler environment model seems to be suited for the task given. Usage of great number of images in segmentation is very time consuming and it was a barrier for delivering better results on this problem.

• Keywords: Genetski algoritam; Genetsko programiranje; Detekcija jezgri stanica; Biomedicina; Segmentacija slike; Morfološke transformacije
• Keywords (english): Genetic algorithm; Genetic programming; Nucleus detection; Biomedicine; Image segmentation; Morphology transformations
• Language: croatian
• URN:NBN: urn:nbn:hr:168:101378
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-03-13 22:48:58