Computational histology image analysis of human cerebral cortex cytoarchitectonics

doctoral thesis

• Title (english): Računalna analiza histoloških slika citoarhitektonike ljudske moždane kore
• Author: Andrija Štajduhar; MBZ: -333878
• Mentor: Sven Lončarić (mentor)
• Mentor: Miloš Judaš (komentor)
• Committee member: Sven Lončarić (predsjednik povjerenstva)
• Committee member: Miloš Judaš (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; (Department of Electronic Systems and Information Processing); Zagreb
• Defense date and country: 2019, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing; Data Processing
• Universal decimal classification (UDC): 621.3 - Electrical engineering

Abstract

The structure of the cerebral cortex is defined by organization and distribution of neurons in the tissue. Today, most investigations in brain cytoarchitectonics are not automatized and require significant involvement of the researcher, which is very time consuming and may introduce significant bias. This type of analysis may not be accurate enough or reveal enough information about the structure of the tissue. Brain tissue forms complex structures whose organization and distribution may be reliably revealed only using systematic computational approach. The goal of this dissertation is development of novel computational methods for automatic and objective analysis of histological images of human brain suitable for advanced and fast analysis of brain cytoarchitectonics and laminar structure. The methods were developed aiming for a completely data-driven approach, with as little assumptions about the data as possible. Quantitative analysis of neuronal distribution and exploration of underlying principles of variation that facilitate parcellation of cortical layers resulted in novel framework and methodology for cellular-level investigations of brain tissue.

Abstract (english)

Strukturu moždane kore definira organizacija i raspodijela neurona u tkivu mozga. Većina današnjih istraživanja u citoarhitektonici mozga nije automatizirana i zahtijeva značajnu uključenost istraživača i mnogo vremena, a što može uvesti i veliku pristranost. Ova vrsta analize često nije dovoljno precizna i ne otkriva dovoljno informacija o strukturi tkiva. Moždano tkivo tvori složene strukture čija se organizacija i distribucija mogu pouzdano otkriti samo uz primjenu sustavnog računarskog pristupa.Cilj ove disertacije je razvoj novih računalnih metoda za automatsku i objektivnu analizu histoloških slika ljudskog mozga pogodne za naprednu i brzu analizu citoarhitektonike mozga i laminarne strukture. Metode su razvijene pomoću pristupa vođenog podacima, uz što manje pretpostavki o samim podacima. Kvantitativna analiza raspodjele neurona i istraživanje temeljnih načela varijacije u tkivu koje uzrokuju parcelaciju kortikalnih slojeva rezultirala je novom znanstvenom metodologijom za istraživanje tkiva mozga na staničnoj razini.

• Keywords: brain cytoarchitecture; partial differential equations; digital image processing; machine learning
• Keywords (english): Citoarhitektonika mozga; parcijalne diferencijalne jednadžbe; obrada digitalne slike; strojno učenje
• Language: english
• URN:NBN: urn:nbn:hr:168:462227
• Study programme: Title: Electrical Engineering and Computing; Study programme type: university; Study level: postgraduate; Academic / professional title: Doktor znanosti elektrotehnike i računarstva (Doktor znanosti elektrotehnike i računarstva)
• Type of resource: Text
• Extent: 120 str. ; 30 cm
• File origin: Born digital
• Access conditions: Open access
• Repository: FER Repository
• Created on: 2020-01-20 10:34:28