UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Computer aided detection of natural reference markers in serial radiographs of growing bone

Springate, Stephen David; (1993) Computer aided detection of natural reference markers in serial radiographs of growing bone. Doctoral thesis (Ph.D.), University College London (United Kingdom). Green open access

[thumbnail of Computer_aided_detection_of_na.pdf] Text
Computer_aided_detection_of_na.pdf

Download (16MB)

Abstract

This thesis describes the theoretical and experimental development and testing of a computer-based image processing system for the detection of stable structural features in serial radiographs of growing bone. The study is divided into three parts. First, a simple theoretical model for the detection of stable structures in a radiographic sequence was developed together with a more detailed statistical model of the processes which hinder detection. Analysis of the models indicated that the processing procedures needed to perform the detection were those described by the so-called "matched filter equation". Secondly, the assumptions of the model were tested in a series of experiments using serial radiographs of 24 children with implanted tantalum markers, drawn from the files of the University of Washington, USA. Information gained from these experiments prompted changes to the detection method to allow the radiographic data to comply with these assumptions and provide robustness against image noise. The method was implemented as a semi-automatic image processing system on a Sun 3/160 computer. The system was tested in a further series of experiments using a second sample of radiographs of 28 children with implanted markers and with radiographs of dried human skulls. The results of these experiments indicated that the system could detect persistent structural features whose positions were consistent with the stable tantalum markers and that the system was robust against minor changes in projection between radiographs. Finally, the detection system was applied to the investigation of mandibular incisor eruption from intra-alveolar to pubertal eruption stages using serial lateral cephalometric radiographs of 11 children. This investigation revealed two new features of incisor eruption: first, the eruption paths deviated lingually at, or just prior to, alveolar emergence; and second, post-emergence eruption was characterised by a labio-lingual oscillation of the crowns of the incisors.

Type: Thesis (Doctoral)
Qualification: Ph.D.
Title: Computer aided detection of natural reference markers in serial radiographs of growing bone
Open access status: An open access version is available from UCL Discovery
Language: English
Additional information: Thesis digitised by ProQuest.
Keywords: (UMI)AAIU057119; Biological sciences
URI: https://discovery.ucl.ac.uk/id/eprint/10098013
Downloads since deposit
39Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item