Use this URL to cite or link to this record in EThOS:
Title: Smoothness-guided 3-D reconstruction for 2-D histological images
Author: Cifor, Rada Amalia
ISNI:       0000 0004 2709 2689
Awarding Body: University of Nottingham
Current Institution: University of Nottingham
Date of Award: 2010
Availability of Full Text:
Access from EThOS:
The microscopic observation of thin sections of anatomical tissue provides knowledge about its molecular and cellular constituents, which is crucial in identifying pathologies, understanding the structure and function of internal organs and for the construction of anatomical atlases. The digitization of these sections yields two dimensional (2-D) images which provide rich anatomical and functional detail at both microscopic and macroscopic level. While the spatial resolution, contrast and specificity of these images continue to outperform the classic three dimensional (3-D) imaging modalities, such as magnetic resonance imaging, their quality and, crucially, quantitative analysis is still limited. The reason is that the organs or anatomical structures of interest are inherently 3-D objects and the analysis of their shape, the computation of their volume, or the comparison of their characteristics across individuals cannot be accurately performed on the basis of 2-D sections alone. Therefore, 3-D volume reconstruction from 2-D histological images usually constitutes a first step in the morphological analysis of the structures imaged by histology. Yet, the loss of 3-D spatial alignment together with the numerous artefacts occurring in the 2-D image acquisition process make reconstruction a difficult task. The work presented in this thesis is based on the observation that the quality of reconstructed histological volumes is usually assessed by considering the smoothness of some reconstructed structures of interest. Our research has two novel contributions: (1) two 3-D reconstruction methods for 2-D histological images which use smoothness as a means to drive the reconstruction process itself; (2) a quantitative measure of smoothness to assess the quality of reconstructed volumes. We apply the reconstruction techniques to various datasets of both synthetic and real histological images. The qualitative visual inspection of the reconstructed volumes is complemented with the quantitative measurements of smoothness with excellent agreement. We also perform a robustness analysis of the proposed reconstruction methods where we evaluate their behaviour in the presence of a variety of geometrical perturbations and typical histological
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available