with clinical and radiological patterns observed in vivo. Methods and Re- sults: Immunohistochemical staining of post-mortem brain tissue reveals two major patterns based on the presence of interneuronal inclusions. The first pattern is a build up of tau either in the form of classic Pick bodies, or diffuse neuronal and glial typically seen in patients with mutations of the tau gene. The second major pattern consists of TDP-43 inclusions found in those with MND associated FTD and more recently with patients with pro- granulin gene mutations. Up to a third of patients have a positive family his- tory and two major causative gene mutations occur on chromosome 17: the microtubule associated protein tau (MAPT) gene and the progranulin gene. Familial motor neuron disease with FTD has been linked to chromosome 9 although no gene is yet identified. Three main variants are recognised. Pa- tients with the behavioural variant present with insidiously progressive changes in personality and behaviour. There is impaired judgement, a lack of initiation and apathy. Social skills deteriorate and there can be socially in- appropriate behaviour, disinhibition, obsessive–compulsive and ritualized behaviour. Emotional labiality and mood swings are seen, but psychotic phe- nomena are rare. Simple bedside cognitive screening tests such as the Mini- Mental State Examination (MMSE:) are insensitive. Imaging reveals changes in the orbital and mesial frontal cortex and the pathology underlying is very variable. In semantic dementia there is a progressive loss in concep- tual knowledge causing anomia and impaired comprehension of words, ob- jects, or faces. Patients are unable to understand less frequent words and fail on a range of semantically based tasks. Repetition of words is normal even though patients are unaware of their meaning. Imaging reveals consis- tent anterior temporal atrophy and the underlying pathology is TDP-43 pos- itive. In progressive non-fluent aphasia there is a gradual loss of expressive language abilities with impairments in the phonological and grammatical aspects of language production. Repetition of multisyllabic words and phrases is impaired but, in contrast to semantic dementia, word comprehen- sion is well preserved. Imaging if often normal and the pathology is typi- cally tau-positive. Conclusions: The diagnosis of FTD is based on the clinical, neuropsychological, and imaging assessments. There is emerging evidence of some correspondence between clinical phenotype and underly- ing pathology which has implications for counseling and for prognosis. 17 ANDI – THE AUSTRALIAN NORMATIVE AND DEMENTIA IMAGING COLLABORATIVE NETWORK: SOFTWARE TOOLS FOR LARGE DATABASES A. Janke 1 , A. Rodell 1 , M. Budge 1 , 1 Dementia Collaborative Research Centre: Prevention, Risk Reduction and Early Detection, Australian National University, Canberra, ACT, Australia Background: Neuroimaging is one of the most utiltised tools in the iden- tification of the dementias. There have been great advances in understand- ing what we now know as "normal" ageing with respects to cortical atrophy and change. Our understanding of the processes that are involved is increasing and yet there is still no conclusive diagnostic test for the de- mentias. There is promise however that the aggregation of imaging, clin- ical phenotyping and possibly genetic information will enable such a test. Methods: The ANDI project aims to generate a database of sufficient size (n>10,000) from existing clinical data sources that will assist with both ed- ucation regarding the dementias via exemplars and staging of dementia in a novel subject. In order to analyse and aggregate large amounts of data a lot of automated pre-processing is needed. These include: robust image re-orientation, artifact detection, non-uniformity correction (MRI only), registration and identification of previous subject data. Novel images can then be compared with age, education and sex-matched normative models such that the individual patients’ data can be staged with respect to com- munity norms in a clinic. Results: The image re-orientation, non-unifor- mity and model generation techniques have been developed. Work is nearing completion for artifact detection. Conclusions: The field of diag- nostic Neuroimaging and Brain Mapping could be greatly advanced by showing a direct link between research in Brain Imaging of the dementias and the resulting impact on Community Health that results from better im- aging techniques being available to clinical practice. 18 BRINGING OUT THE HUMAN ESSENCE’ IN PEOPLE WITH DEMENTIA H. Lee 1,2 , A. Orb 1 , 1 The Society for the Arts in Dementia Care (Australia) Inc; 2 Dementia Consultant Aims: The Australian government recognizes the need to consider new psy- chosocial models in dementia care, particularly in relation to quality of life and wellbeing. The Spark of Life club program is a small group activity that incorporates a person-centred approach and is conducted by two trained facil- itators. It is used internationally for those with early to advanced dementia. The aim of the study was to examine carers’ and families’ perceptions of the impact of the Spark of Life club program on the personal and emotional wellbeing of people with dementia. Method: 17 semi-structured interviews were conducted with carers and families from a facility implementing this program for people with moderate dementia. Interviews were taped and transcribed verbatim and content analysis was used to analyse the data. Results: Four main themes emerged from the data: 1. Re igniting the human essence in which participants indicated improvement in club members’ vitality and interest in life, memory, confidence, ability to communicate and socialise, 2. Developing a sense of cre- ativity where club members expressed enjoyment through music, humour, and story telling. 3. Being in their shoes illustrated how staff and relatives became aware of club members’ potential for improvement and learning, 4. Enabling success highlighted the importance of the staff attributes and resources avail- able that contributed to the positive outcomes of the program. Conclusions: The study concludes that this program has important therapeutic outcomes both for people with dementia and their families. Moreover, the research indi- cates that valuable benefits for dementia practitioners are to be gained training in the Spark of Life. Further research is recommended. 19 USE OF THE INTRANET IN DEMENTIA EDUCATION IN ACUTE CARE D. Lie 1 , S. Austin 2 , 1 Princess Alexandra Hospital, Wooloongabba, QLD, Australia; 2 University of Queensland, Wooloongabba, QLD, Australia Background: In Australia, there are 227,300 individuals with dementia with 57,000 expected to be newly diagnosed in 2008. The average hospital length of stay is 26.4 days of separation for individuals with a principal diagnosis of dementia. Complexities of care around behavioral and psy- chological symptoms of dementia (BPSD) are highly prevalent in the acute care setting; including falls, wandering, aggression and refusal of care. The purpose of this study is to investigate the use of web-based material incorporating the intranet of a metropolitan teaching hospital. There is lim- ited information about effectiveness of web-based material and log file analysis in the health literature. This software allows reporting of web use patterns, frequency, duration and time of use. Methods: A website has been established on the intranet of a metropolitan teaching hospital; July 1 st 2008–June 30 th 2009. The website incorporates multimedia and text files; including short vignettes about management of behaviors of con- cern. Promotion of the website is; emails to the clinical educators, Nursing Unit Managers, and staff. Results: Logfile analysis will indentify the most common areas of interest and pattern of usage. Staff can download mate- rial onto WMV [Windows Media Player], Div X, MPEG4 and other por- table devices. Conclusions: Web-based education is increasingly used but its utility in acute care settings requires evaluation. This study is innova- tive in focussing on dementia care and in using log-file analysis to provide evidence of utility. 20 HOW TO IMPROVE PROXY MEASUREMENT IN DEMENTIA: SUGGESTIONS FROM THE SCIENTIFIC LITERATURE N. Marosszeky 1 , J. Sansoni 1 , 1 Australian Health Outcomes Collaboration, Centre for Health Service Development, University of Wollongong, Wol- longong, NSW, Australia Background: The progressive deterioration of cognitive and functional abil- ities in people with dementia, limits the applicability of direct questioning or interview methods to assess a person’s symptoms, behaviors and health Abstracts / Alzheimer’s & Dementia 5 (2009) e1–e9 e5