Visual tracking and reconstruction of 3D curves in medical image analysis

doctoral thesis

• Title (croatian): Vizualno pracenje i rekonstrukcija 3D krivulja za analizu medicinskih slika
• Author: Tomislav Petković
• Mentor: Sven Lončarić (mentor)
• Committee member: Zoran Kalafatić (predsjednik povjerenstva)
• Committee member: Sven Lončarić (član povjerenstva)
• Committee member: Srećko Lončarić (član povjerenstva)
• Committee member: Damir Seršić (član povjerenstva)
• Committee member: Davor Petrinović (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; (Department of Electronic Systems and Information Processing); Zagreb
• Defense date and country: 2010-09-24, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing; Data Processing
• Universal decimal classification (UDC): 004 - Computer science and technology. Computing. Data processing

Abstract

Visual tracking of moving objects is one of important open problems in the field of computer vision. Due to complexity instead of general purpose only solutions for well-defined applications are usually considered. Furthermore, visual tracking in the field of bio-medicine is often based on transmissive and projective acquisition model making existing procedures not applicable as the acquisition model is not compatible. Better results can be achieved by utilizing methods adapted to the acquisition model. In this work a modular tracking algorithm to visually track moving objects that are represented by a 3D curve is presented. Described algorithm is adapted to the acquisition model. As the acquisition geometry is projective reconstruction of the position and orientation is also discussed with limitations to reconstruction uniqueness and accuracy given in regard to the number of views. Guide-wire tracking during minimally invasive intravascular interventions is used as an example.

Abstract (croatian)

Vizualno pracenje pokretnih objekata predstavlja jedan od vaznih otvorenih problema racunalnog vida. Zbog slozenosti problema se umjesto opceg konstruiraju algoritmi samo za usko denirane primjene. Nadalje, vizualno pracenje u bio-medicini se cesto temelj na transmisijskom projektivnom snimanju pa postoje ci postupci nisu prikladni jer ne odgovaraju takvom modelu snimanja. Bolji rezultati se mogu ocekivati upotrebom metoda prilagodenih takvom snimanju. U okviru disertacije predlozen je modularni postupak za vizualno pracenje pokretnih objekata koji se mogu priblizno opisati 3D krivuljom. Opisani postupak je dodatno prilagoden transmisijskom projektivnom snimanju. Kako se radi o projektivnom snimanju 3D objekta osim pracenja razmatra se i problem rekonstrukcije polozaja i orijentacije te se daju ogranicenja na jedinstvenost i tocnost rekonstrukcije ovisno o broju dostupnih pogleda. Kao primjer primjene daje se pracenje katetera tijekom minimalno invazivnih vaskularnih zahvata.

• Keywords: 3D curve reconstruction from projections; transmissive imaging; guidewire tracking; guidewire reconstruction
• Keywords (croatian): rekonstrukcija 3D krivulje iz projekcija; transmisijsko snimanje; pracenje katetera; rekonstrukcija katetera
• Language: english
• URN:NBN: urn:nbn:hr:168:047114
• Project: Number: 036-0362214-1989; Title: Inteligentne metode obrade i analize slika; Leader: Sven Lončarić; Jurisdiction: Croatia; Funder: MZOS; Funding stream: ZP
• Study programme: Title: Computer Science; Study programme type: university; Study level: postgraduate; Academic / professional title: Doktor znanosti (Doktor znanosti)
• Catalog URL: http://lib.fer.hr/cgi-bin/koha/opac-detail.pl?biblionumber=34322
• Type of resource: Text
• Extent: 158 str.
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-03-25 12:18:42