Melbourne Dental School - Theses
Permanent URI for this collection
Search Results
1 - 1 of 1
-
ItemApplication of 3D mapping of the face for forensic application using reconstructive anatomical techniques( 2009)The human face has especially important functions in interactions between people whose capacity to recognise other individuals and interpret their emotions is largely an innate skill reinforced by early life experiences. The human face not only plays a key role in everyday communication but it also helps define the individual’s identity because it is highly likely that any particular combination of facial features is unique. The concept of “race” is controversial amongst anthropologists. There are, however, groups of people who share physical characteristics that have evolved in response to long-term prevailing environmental conditions. Similarly, when asked to identify the ancestry of themselves and their family group, people align themselves with others having shared facial characteristics. In the past, defining such characteristics quantitatively has proved to be difficult largely because of shortcomings in measurement techniques. Most of these techniques take measurements directly from the face. More recently, advances in 3D scanning technology has enabled researchers to collect accurate information from the face such as topography, colour and texture indirectly. The first aim of this study was to collect a set of facial images of people identifying themselves as ‘Malays’ who originated from Malaysia and to develop a characteristic adult facial archetype (“average face”) for men and women. Facial images from 50 volunteer participants (25 male and 25 female) with self-declared Malay ethnic background had their faces scanned using the Konica Minolta VIVID 910 3D laser scanner. Data collected were processed through Geomagic® software to produce the final digital images. Male and female Malay facial archetypes were created from these 50 images. Then, using the software developed in-house by Claes (2007), 36 predefined anatomical landmarks were indicated on each individual’s facial image and the corresponding facial archetype, (male or female). This process was repeated 10 times to measure reproducibility of landmark identification. The second aim was to compare the data from individuals to determine variances within the Malay population to their corresponding archetype. Comparisons of these results with data representing Australian Caucasian and Japanese populations were also made. The analysis took several approaches in a logical sequence striving to find the most meaningful way to interpret the data for the purposes of comparing different populations in a quantifiable way. The first analysis used Euclidean Distance Matrix Analysis (EDMA) in which 5 groups were compared: males and females from the Malay population, Malay and Australian Caucasian males, Malay and Japanese males, Malay and Australian Caucasian females and lastly between the Malay and Japanese females. For the Malay cohort the position of landmarks indicated on the facial images of individuals were compared with corresponding landmarks indicated on the corresponding facial archetypes as a measure of intra-operator error and thus provided an indication of the validity of sparse landmark based analysis. The finding that intra-operator error was of the same order as measurable differences between individuals meant that such an approach is practically worthless and so the final analyses were done using automatically generated sets of dense corresponding landmarks on the Malay facial images and the same five groups compared previously. Contrary to what has been believed previously by many other workers, even with implementation of very rigorous internal quality assurance, the results from the analysis of the facial data using EDMA demonstrated that sparse sets of landmarks were not useful to distinguish between populations. This was because of the difficulties in locating landmarks sufficiently accurately and reproducibly. The dense correspondence landmark method was more successful in providing quantifiable differences between the groups that can be easily depicted. This provides the first 3D archetypes of Malay men and women.