Forensic Dentistry has highlighted a pressing need for a better and more precise ageing technique in the identification of adult human remains. This Thesis is a fundamental study of apical third odontoblasts and adjacent dentine [the pulpo-dentinal complex] in vital, disease-free human teeth of various ages using similar carnivore teeth as an animal model. The role of intratubular dentine (ITD) in the ageing of tubular dentine has been questioned. The nature of the organic matrix, authoritatively believed to be non-collagenous and providing the network for apatite mineralisation, offered a vehicle which may involve a component that could act as a marker for ITD development with age. A non-collagenous protein manufactured in the liver, Alpha2HSglycoprotein (α2HS), which was thought to implicate the odontoblasts in its transmission to the dentinal tubule and known to be a part of hard tissue mineralisation, was investigated by immunohistochemical means. Previous monoclonal antibodies could not be reproduced therefore a new immunohistochemical protocol based on polyclonal antibodies was devised. Positive and negative controls were used. The extent of α2HS was quantitatively assessed by a modified Image Pro-plus Programme: SPSS used for data analysis. Correlation of ITD with age using α2HS is discussed together with the involvement of the protein in 'occlusive material' and amorphous pulpal dentine. The age changes in the apical third of the root of vital, disease-free, single-rooted teeth whilst occurring as age advanced may be less obviously correlated to age than previously thought. The use of the extent of ITD as an age determinant was considered not to be proven. The use of the carnivore model was effective and reliable.

This thesis is an investigation into the use of Volume Graphics techniques for the purpose of forensic facial reconstruction. The thesis also examines the rapid creation of a light volume for the accurate rendering of light within a scene. The main achievements of this thesis are: an examination of forensic facial reconstruction techniques, both manual and automatic; identification of weaknesses of these methods and an improved automatic method using a volume graphics approach; a general introduction to the rapidly evolving field of volume graphics; the introduction of a technique for calculating fast, accurate measurements of light intensities inside a volume dataset. This work has demonstrated that it is now feasible to use a computer to reconstruct the face of an unknown person given his/her skull without allowing artistic interpretation. It has also demonstrated that it is computationally feasible to calculate a “light volume” to store light intensities at voxel points within a volume, thus allowing for much faster recreation of shadowing effects e.g. during rendering. Part of this research was presented by the candidate at the 19^th Eurographics UK conference at the University College, London, 2001.

When skeletalised or semi-skeletalised remains are discovered there is often no identifying evidence available. In these cases, facial reconstruction techniques are employed to re-create a pre-mortem likeness of the deceased in the hope of developing new leads towards achieving the positive identification of the deceased. This work is a thorough investigation into the feasibility of using volume graphics techniques to perform the facial reconstruction of skeletal remains. The main contributions are: (i) A critical review of existing methods for facial reconstruction and soft tissue depth measurement. The aim of this study is to determine which factors contribute most to the inaccuracy of current reconstruction methods; (ii) The detailed design of a pipeline for the reconstruction of facial features using volume graphics techniques; (iii) A new approach for extracting corresponding anatomical feature points from two volumetrically-defined skulls; (iv) A new algorithm for the smooth and accurate deformation of volumetric data; (v) The evaluation of the new reconstruction pipeline with real-life skeletal samples. This research clearly illustrates the feasibility and potential usefulness of using volume graphics as the basis for a computer-based facial reconstruction technique. Parts of this research have been presented by the candidate at "Eurographics UK 1996" (Imperial College, London), and "International Workshop on Volume Graphics 1999" (Swansea, United Kingdom). It has also been published in 'Forensic Science International' and "Volume Graphics" (Springer-Verlag, London).

Shoeprints are routinely left at crime scenes and are reported to be present more frequently than fingerprints. It has been reported that 35 percent of crime scenes present shoe marks that can be recovered and used as forensic evidence. During investigations, a scene of crime shoeprint can be matched against a database of known shoeprints in order to identify the brand and the model of the corresponding shoe. TIlls is known as shoeprint classification and is, currently, performed manually or using some semi-automatic systems. These current approaches are time consuming and are not very reliable. Thus, the development of automatic shoeprint classification methods would offer valuable assistance to forensic scientists. TIlls thesis addresses the task of automatic shoeprint classification and its related challenges. This includes the problem of classifying partial, noisy and/or blurred shoeprint images. The issues of invariance to geometric distortions, e.g. translations and rotations, as well as rapid classification are also considered. The thesis proposes a number of different ideas and methods for the automatic classification of distorted shoeprint images including the use of Fourier-Mellin transform, modified phase-only correlation and two-dimensional advanced correlation filters. It also investigates the use of multiple one-dimensional correlation filters and classifier combination techniques, such as algebraic rules, Decision Templates and Support Vector Machine based combiners. The experimental results suggest that the investigated correlation-based methods can offer high accuracies when classifying low quality shoeprint images while providing tolerance to geometric distortions.

No description available.

Raman spectroscopy is currently the subject of much interest from the forensic science community due to its advantageous properties. In this Thesis, the technique is employed in the examination of several evidence types such as multi-layer white paint, inks, a new class of drug-of- abuse and explosives. For the multi-layer white paint chip, layers could be delineated on the basis of the changing chemical composition between paints in different layers. For the inks, Raman spectroscopy provided good discrimination and in some cases, better than the conventional techniques employed. This enabled construction of a flow chart was to direct the forensic document examiner to the most efficient techniques to use." Legal highs" are a new class of drug-of-abuse posing considerable problems for the criminal justice system, in particular the l3-ketophenethylamines. Within forensic casework, significant volumes of these materials are still encountered. Therefore, data on the Raman studies of a series of reference standards and seized samples are presented and demonstrate that the technique provides a rapid and effective method to identify these materials and any bulking materials present. Finally, data are presented on two compounds used in the manufacture of improvised explosive devices, ammonium nitrate based fertiliser and hydrogen peroxide. Using Raman spectroscopy, information on the composition of the fertiliser encountered within these devices could be rapidly obtained. A small number of seized samples were examined and it was demonstrated that there was significant variance between them that allowed discrimination. Peroxide based explosives are currently one of the explosives of choice for terrorists and here, a method to semi-quantitatively identify the concentration of hydrogen peroxide using a handheld spectrometer. This thesis demonstrated the significant benefit that could be obtained by implementing Raman spectroscopy in the standard operating procedure for several evidence types.

No description available.

No description available.

An electroimmunodiffusion technique, which enables precise measurements of the concentrations of some plasma proteins in blood stain extracts to be made, has been developed. It is possible that if a number of plasma proteins, with high discriminating power were found to be stable in blood stains, and were measured by this technique, that the information so obtained could be of value for the forensic examination of blood stains. The tandem antigen–antibody crossed electrophoretic technique has also been examined, and its application to this problem assessed. Nine plasma proteins have been examined, and some of them found to be stable in blood stains. Additional immunological techniques l1ave been employed in an attempt to establish the nature of the changes which the unstable proteins underwent. Seven of the proteins were quantitated in samples provided in a number of blind trials, in order to test the feasibility of a plasma profiling system for matching blood stains. Finally, the advantages and disadvantages of a plasma profiling system over existing methods of blood stain analysis have been discussed.

The accurate estimation of age at death of human skeletal remains is an important aspect of many disciplines including archaeology, palaeopathology and medicine. The aim of this project was to investigate five aspects of the endocranial surface of the skull, mid-parietal thickness, hyperostosis frontalis interna (HFI), cranial suture fusion, vascular grooves and arachnoid granulations and to evaluate their relationship with age. Four samples (total 697 skeletons) were used: (1) A modern post-mortem population; (2) an early 20th century American anthropological collection; (3) A 17th-19th century French/English archaeological sample; (4) An English 10th-19th century archaeological sample. Three of these samples were of known age at death. Direct measurement, recording and X-ray techniques were used to evaluate the relationship between specific bone changes and age. A measurable, but weak relationship with age was observed in skull mid-parietal thickness, HIF, rate of endocranial and ectocranial suture fusion and the cross sectional profile of meningeal vessel grooves. A new technique for the diagnosis of HFI was proposed, based on standard x-ray score. Use of this method on the four samples showed that HFI is not increasing in prevalence with modernity, as previously thought. A strong relationship was observed between arachnoid granulation pit counts and age. A new method for estimating age at death was proposed and tested on two populations of known age. This new method was as accurate in estimating age at death as two of the most popular techniques currently in use. The method has less bias in ageing, can be applied to both sexes with equal accuracy and has no maximum age cut-off point.