Avtomatska segmentacija belih madežev na gladkih zobnih ploskvah

Kozjek, Klemen (author), Šajn, Luka (mentor)

Vse pogosteje se za medicinske namene uporablja računalniške tehnologije, saj te predstavljajo pomemben gradnik sodobnega zdravstva. Po snetju nesnemljivega ortodontskega aparata se pogosto pojavijo začetki demineralizacije, ki se kažejo kot bele ploskve na gladkih površinah zobne sklenine. V magistrskem delu razvijemo prototip sistema za avtomatsko segmentacijo zob in belih madežev, ki lahko pripomore k bolj natančnemu in objektivnemu spremljanju zdravljenja. Pri razvoju smo uporabili različne tehnike za obdelavo slik in algoritmov za segmentacijo. Razvit sistem smo ovrednotili na podatkovni zbirki, katero smo pripravili iz izbranih slik kliničnih pregledov. Rezultati so pokazali, da sistem deluje in ima še veliko prostora za nadaljnje delo ter izboljšave.

• Language: Slovenian
• Keywords: računalniški vid, obdelava slik, beli madeži, segmentacija slik, rast regij, segmentacija z grafi
• Work type: Master's thesis/paper
• Organization: FRI - Faculty of Computer and Information Science
• Year: 2018
• PID: 20.500.12556/RUL-103815
• Publication date in RUL: 26.09.2018

KOZJEK, Klemen, 2018, Avtomatska segmentacija belih madežev na gladkih zobnih ploskvah [online]. Master’s thesis. [Accessed 23 March 2025]. Retrieved from: https://repozitorij.uni-lj.si/IzpisGradiva.php?lang=eng&id=103815

Secondary language

• Language: English
• Title: Automatic segmentation of white spot lesions on smooth tooth surfaces

Abstract

Computer technologies are ubiquitous and in the recent times, these technologies are penetrating more into the field of medicine, where they play a vital role in modern healthcare. In this master's thesis, we are solving a problem, which dentists typically encounter during teeth alignment treatment. At the end of the treatment, when permanent orthodontic braces are removed, the initial phase of tooth demineralization often appears as white spot lesions on the smooth surfaces of a tooth. We developed a prototype for automatic segmentation of teeth and white spot lesions, which may contribute to a more accurate and objective way of treatment monitoring. In the process of development, we used various image processing techniques and image segmentation algorithms. The developed prototype was evaluated against a database, which we built from the selected images of clinical examinations. The prototype showed promising results with a lot of potential for improvements and future work.

• Keywords: computer vision, image processing, white lesions, image segmentation, region growing, segmentation with graphs