Frost S.,CSIRO |
Frost S.,Australian alth Research Center |
Frost S.,University of Western Australia |
Kanagasingam Y.,CSIRO |
And 20 more authors.
Translational Psychiatry | Year: 2013
The earliest detectable change in Alzheimer's disease (AD) is the buildup of amyloid plaque in the brain. Early detection of AD, prior to irreversible neurological damage, is important for the efficacy of current interventions as well as for the development of new treatments. Although PiB-PET imaging and CSF amyloid are the gold standards for early AD diagnosis, there are practical limitations for population screening. AD-related pathology occurs primarily in the brain, but some of the hallmarks of the disease have also been shown to occur in other tissues, including the retina, which is more accessible for imaging. Retinal vascular changes and degeneration have previously been reported in AD using optical coherence tomography and laser Doppler techniques. This report presents results from analysis of retinal photographs from AD and healthy control participants from the Australian Imaging, Biomarkers and Lifestyle (AIBL) Flagship Study of Ageing. This is the first study to investigate retinal blood vessel changes with respect to amyloid plaque burden in the brain. We demonstrate relationships between retinal vascular parameters, neocortical brain amyloid plaque burden and AD. A number of RVPs were found to be different in AD. Two of these RVPs, venular branching asymmetry factor and arteriolar length-to-diameter ratio, were also higher in healthy individuals with high plaque burden (P=0.01 and P=0.02 respectively, after false discovery rate adjustment). Retinal photographic analysis shows potential as an adjunct for early detection of AD or monitoring of AD-progression or response to treatments. © 2013 Macmillan Publishers Limited All rights reserved.
D'Souza M.,CSIRO |
Wark T.,CSIRO |
Karunanithi M.,Australian alth Research Center |
Ros M.,University of Wollongong
Pervasive and Mobile Computing | Year: 2013
We present the development and evaluation of a realtime indoor localisation system for tracking people. Our aim was to track a person's indoor position using dead-reckoning, while limiting position error without depending on extensive wireless network infrastructure. The Indoor People Tracker used wearable motion sensors, a floor-plan map and a limited wireless sensor network for proximity ranging. We evaluated how the position accuracy of the Indoor People Tracker was affected by floor-plan map features, wireless proximity range and motion information. The advantage of the Indoor People Tracker was found; it was able to achieve accurate position resolution with minimal error, while not depending on wireless proximity.© 2012 Elsevier B.V. All rights reserved.
Wang Y.,University of New South Wales |
Wang Y.,Australian alth Research Center |
Cheema M.A.,University of New South Wales |
Lin X.,University of New South Wales |
And 2 more authors.
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2013
Manifold Ranking (MR) is one of the most popular graph-based ranking methods and has been widely used for information retrieval. Due to its ability to capture the geometric structure of the image set, it has been successfully used for image retrieval. The existing approaches that use manifold ranking rely only on a single image manifold. However, such methods may not fully discover the geometric structure of the image set and may lead to poor precision results. Motivated by this, we propose a novel method named Multi-Manifold Ranking (MMR) which embeds multiple image manifolds each constructed using a different image feature. We propose a novel cost function that is minimized to obtain the ranking scores of the images. Our proposed multi-manifold ranking has a better ability to explore the geometric structure of image set as demonstrated by our experiments. Furthermore, to improve the efficiency of MMR, a specific graph called anchor graph is incorporated into MMR. The extensive experiments on real world image databases demonstrate that MMR outperforms existing manifold ranking based methods in terms of quality and has comparable running time to the fastest MR algorithm. © Springer-Verlag 2013.
Wang Y.,University of New South Wales |
Wang Y.,Australian alth Research Center |
Lin X.,University of New South Wales |
Zhang Q.,University of New South Wales |
And 2 more authors.
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2014
In this paper, we develop a novel paradigm, namely hypergraph shift, to find robust graph modes by probabilistic voting strategy, which are semantically sound besides the self-cohesiveness requirement in forming graph modes. Unlike the existing techniques to seek graph modes by shifting vertices based on pair-wise edges (i.e, an edge with 2 ends), our paradigm is based on shifting high-order edges (hyperedges) to deliver graph modes. Specifically, we convert the problem of seeking graph modes as the problem of seeking maximizers of a novel objective function with the aim to generate good graph modes based on sifting edges in hypergraphs. As a result, the generated graph modes based on dense subhypergraphs may more accurately capture the object semantics besides the self-cohesiveness requirement. We also formally prove that our technique is always convergent. Extensive empirical studies on synthetic and real world data sets are conducted on clustering and graph matching. They demonstrate that our techniques significantly outperform the existing techniques. © 2014 Springer International Publishing.
Squitti R.,AFaR Ospedale Fatebenefratelli |
Squitti R.,Laboratory of Neurodegeneration |
Simonelli I.,IRCCS San Raffaele Pisana |
Simonelli I.,AFaR Fatebenefratelli Hospital San Giovanni Calibita |
And 7 more authors.
Journal of Alzheimer's Disease | Year: 2014
The fraction of copper not bound to ceruloplasmin seems altered in Alzheimer's disease (AD). We have addressed this notion evaluating all the studies carried out from 1996 until March 2013 by means of meta-analysis. We performed our analysis on diverse indices evaluating the relationship between copper and ceruloplasmin in general circulation, namely 'Non-Cp copper', '% Non-Cp copper', and 'Adjusted copper'. For Non-Cp copper and % Non-Cp copper, the correct stoichiometry between copper and ceruloplasmin (6-8 atoms of copper for each ceruloplasmin molecule) in healthy controls has been adopted as criterion for the study to be included in the meta-analysis evaluating data with the canonic Walshe's formula for Non-Cp copper. Copper to ceruloplasmin ratio (Cu:Cp), which is an internal quality control check for ceruloplasmin calibration, was used as an index of the actual stoichiometry in the specimens. Adjusted (Adj-Cp) copper, even though less reliable, was calculated, allowing the evaluation of all the studies selected. An additional meta-analysis of systemic total copper was re-calculated accounting for all the studies carried out from 1983 to March 2013. Ten studies were analyzed in the meta-analysis for Non-Cp copper and % Non-Cp copper reaching a pooled total of 599 AD subjects and 867 controls. For Adj-Cp copper, 14 studies were analyzed with a pooled total of 879 AD and 1,712 controls. 27 studies were considered for systemic total copper meta-analysis, with a pooled total of 1,393 AD and 2,159 controls. All the copper indices analyzed were significantly higher in AD subjects compared to healthy controls. © 2014 - IOS Press and the authors. All rights reserved.
Leroux H.,Australian alth Research Center |
McBride S.,Australian alth Research Center |
Gibson S.,Australian alth Research Center
Studies in Health Technology and Informatics | Year: 2011
Clinical research studies offer many challenges for their supporting information systems. AIBL assembled 1112 participants who volunteered crucial information for a comprehensive study on neurodegenerative diseases. This paper discusses the shortcomings of the clinical trial management system chosen to record the results of the study. A set of guidelines was devised and a critique of five systems ensued. OpenClinica was selected as the most appropriate option. The main contribution of this paper is: (i) proposing a set of guidelines to determine the appropriateness of Clinical Trial Management Systems (CTMS) solution; (ii) providing a brief critique of existing commercial and open-sourced CTMS; and (iii) alluding to some data migration issues and providing cues on how to address them. We conclude that open-source CTMS are viable alternatives to the more expensive commercial systems to conduct, record and manage clinical studies. © 2011 The authors and IOS Press. All rights reserved.
Cheung V.H.,University of Queensland |
Cheung V.H.,Australian alth Research Center |
Gray L.,University of Queensland |
Karunanithi M.,Australian alth Research Center
Archives of Physical Medicine and Rehabilitation | Year: 2011
Objectives: To review studies that used accelerometers to classify human movements and to appraise their potential to determine the activities of older patients in hospital settings. Data Sources: MEDLINE, CINAHL, and Web of Science electronic databases. A search constraint of articles published in English language between January 1980 and March 2010 was applied. Study Selection: All studies that validated the use of accelerometers to classify human postural movements and mobility were included. Studies included participants from any age group. All types of accelerometers were included. Outcome measures criteria explored within the studies were comparisons of derived classifications of postural movements and mobility against those made by using observations. Based on these criteria, 54 studies were selected for detailed review from 526 initially identified studies. Data Extraction: Data were extracted by the first author and included characteristics of study participants, accelerometers used, body positions of device attachment, study setting, duration, methods, results, and limitations of the validation studies. Data Synthesis: The accelerometer-based monitoring technique was investigated predominantly on a small sample of healthy adult participants in a laboratory setting. Most studies applied multiple accelerometers on the sternum, wrists, thighs, and shanks of participants. Most studies collected validation data while participants performed a predefined standardized activity protocol. Conclusions: Accelerometer devices have the potential to monitor human movements continuously to determine postural movements and mobility for the assessment of functional ability. Future studies should focus on long-term monitoring of free daily activity of a large sample of mobility-impaired or older hospitalized patients, who are at risk for functional decline. Use of a single waist-mounted triaxial accelerometer would be the most practical and useful option. © 2011 by the American Congress of Rehabilitation Medicine.
Butt L.,Australian alth Research Center |
Zuccon G.,Australian alth Research Center |
Nguyen A.,Australian alth Research Center |
Bergheim A.,Cancer Institute NSW |
Grayson N.,Cancer Institute NSW
CEUR Workshop Proceedings | Year: 2012
The timely notification of cancer cases is crucial for can- cer monitoring and prevention. However, the abstraction and classification of cancer from the free-text of pathology reports and other relevant documents, such as death certificates, are complex and time-consuming activities. In this paper we investigate approaches for the automatic de- tection of cases where the cause of death is a notifiable cancer from free-text death certificates supplied to Cancer Registries. A number of machine learning classifiers were investigated. A large set of features were also extracted using natural language techniques and the Medtex toolkit; features include stemmed words, bi-grams, and concepts from the SNOMED CT medical terminology. The investigated approaches were found to be very effective in identifying death certificates where the cause of death was a notifiable cancer. Best performance was achieved by a Support Vector Machine (SVM) classifier with an overall F-measure of 0.9647 when evaluated on a set of 5,000 free-text death certificates. This classifier considers as features stemmed token bigrams and information from SNOMED CT concepts filtered by morphological abnormalities and disorders. However, our analysis shows that it is the selection of features that most inuences the performance of the classifiers rather than the type of classifier or the feature weighting schema. Specifically, we found that stemmed token bigrams with or without SNOMED CT concepts are the most effective feature. In addition, the combination of token bi- grams and SNOMED CT information was found to yield the best overall performance.
Coles T.R.,Australian alth Research Center |
Dumas C.,Australian alth Research Center
Ergonomics in Design | Year: 2013
The authors report their experience from various task analysis projects in which users have been observed in situ. Solutions for simultaneous video recording, often needed to adequately observe and analyze a workspace, are discussed. Of the various analog- and digital-based solutions, the authors deem a low-cost solution using entry-level computer hardware to produce sufficiently high-fidelity feedback for most task analysis purposes. Such systems can be acquired and set up by novice computer users for a fraction of the cost of broadcast video systems. A case study demonstrates the application of one such inexpensive solution. © 2013 Human Factors and Ergonomics Society.
Hansen D.,Australian alth Research Center
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2012
Health systems around the world are increasing the adoption of electronic health records to improve access to patient data for administrative, treatment and research purposes. The complexity of the data which is being stored in these electronic health records is providing a number of challenges for health information systems. These challenges cover the capture of standardised data in an individual health information system, extracting necessary information from medical narratives, through to the processing of complex data including physiological, biomedical imaging and genomic data. However the largest challenges are in the interchange of the data and having health information systems use the data to provide intelligent decision support. © 2012 Springer-Verlag.