Platformski neovisna udaljena vizualizacija volumnih medicinskih podataka

doctoral thesis

• Title (english): Cross-platform remote visualization and volume rendering of medical data
• Author: Krešimir Jozić; MBZ: 41674
• Mentor: Željka Mihajlović (mentor)
• Committee member: Alan Jović (predsjednik povjerenstva)
• Committee member: Tomislav Hrkać (član povjerenstva)
• Committee member: Goran Krstačić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; (Department of Electronics, Microelectronics, Computer and Intelligent Systems); Zagreb
• Defense date and country: 2022-07-14, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing
• Universal decimal classification (UDC): 004 - Computer science and technology. Computing. Data processing

Abstract

Radiološke snimke pohranjene u DICOM obliku zbog svoje veličine mogu predstavljati problem korisnicima jer su računala na kojima se pregledavaju pod-optimalna. Učitavanje takvih snimki, kao i generiranih volumnih podataka, traje dugo što demotivira korisnike. U slučaju da se snimke ne pregledavaju u ustanovi u kojoj su nastale, prenose se na optičkim ili USB medijima. Kako se snimke obično pregledavaju samo jednom, optički mediji se odbacuju, a to kao posljedicu ima onečišćenje okoliša. Ako se snimke prenose preko Interneta i dalje ostaje problem pod-optimalnih računala ali i cijene usluge korištenja Interneta. Situacija je dodatno pogoršana COVID-19 pandemijom što otežava prijenos medija. Osim toga, rad od kuće je uzrokovao povećanu potražnju za poluvodičima, što je dovelo do nestašice i porasta cijene poluvodiča na tržištu. Disertacija je usmjerena na rješavanje problema tako da predstavi rješenje za daljinsko iscrtavanje radioloških snimki. Rješenje je predstavljeno kroz tri dijela: izradu modela programske potpore, arhitekturu i tehničku analizu ostvarenog sustava. Provedeno istraživanje pokazuje da je moguće ostvariti prikazivanje pojedinačnih slojeva radiološke snimke gotovo trenutno što korisnicima olakšava rad te ubrzava postavljanje dijagnoze.

Abstract (english)

Radiological images stored in DICOM format, due to their size, can be a problem for users because the computers on which they are viewed are sub-optimal. Loading such recordings, as well as generating volume data, takes a long time, which demotivates users. In case the recordings are not reviewed in the institution where they were created, they are transferred on optical or USB media. As recordings are usually viewed only once, the optical media is discarded, resulting in environmental pollution. If the recordings are transferred via the Internet, the problem of sub-optimal computers and the price of the Internet usage service remains. The situation is further aggravated by the COVID-19 pandemic, which makes media transmission difficult. In addition, working from home has caused an increased demand for semiconductors, leading to a shortage and an increase in the price of semiconductors in the market. The dissertation is aimed at solving the problem by presenting a solution for the remote rendering of radiological images. The solution is presented in three parts: the creation of a program support model, architecture, and technical analysis of the realized system. The conducted research shows that it is possible to display the individual layers of the radiological image almost instantly, which makes it easier for users to work and speeds up the diagnosis.

• Keywords: radiološke snimke; grafičko sklopovlje; iscrtavanje volumena; daljinsko iscrtavanje
• Keywords (english): radiological images; graphic hardware; volume rendering; remote rendering
• Language: croatian
• URN:NBN: urn:nbn:hr:168:296996
• Promotion: 2022
• Study programme: Title: Computer Science; Study programme type: university; Study level: postgraduate; Academic / professional title: Doktor znanosti (Doktor znanosti)
• Type of resource: Text
• Extent: viii, ii, 78 str : graf. prikazi
• File origin: Born digital
• Access conditions: Open access
• Repository: FER Repository
• Created on: 2022-09-05 08:12:17