By Topic

Intelligent Information Systems Conference, The Seventh Australian and New Zealand 2001

Date 18-21 Nov. 2001

Filter Results

Displaying Results 1 - 25 of 83
  • The management of medical technology

    Publication Year: 2001 , Page(s): 367 - 372
    Cited by:  Papers (2)
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (501 KB) |  | HTML iconHTML  

    Medical technology is a valuable asset that is strategically important to the operational efficiency of the Australian health care industry. This paper looks at the Australian care health industry and suggests the need for a more structured approach to medical technology strategic planning at a national, state, and hospital level. Strategic planning frameworks are considered and data relating to a sample of Australian hospitals is discussed. The paper deals with a wide range of issues that impact on medical technology strategic planning including cost of ownership, access and equity, remaining life of technology, financing options, and centralised vs local planning. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Data representation influences protein secondary structure prediction using artificial neural networks

    Publication Year: 2001 , Page(s): 411 - 415
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (366 KB) |  | HTML iconHTML  

    Artificial Neural Networks (ANN) have been used very successfully for a number of classification problems in the molecular biology field. Protein secondary structure prediction is one of the oldest and best defined of these classification problems. Yet despite the considerable amount of work conducted in this field there still remain a number of fundamental computational issues that have not been thoroughly investigated, if considered at all. One important issue is identifying an appropriate data representation for input into the ANN. In this paper, we have investigated a range of new encoding schemes and evaluated their performance using recently introduced evaluation criterion. We have done this by preserving the redundant information of DNA codons that is lost when they are translated into amino acids. Interestingly, with our new data representation, the β-strand prediction performance was consistently higher (14% improvement) over the accuracy of the ANNs trained when the conventional representation was used. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Shape understanding: knowledge generation and learning

    Publication Year: 2001 , Page(s): 189 - 195
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (527 KB) |  | HTML iconHTML  

    Image analysis and recognition applied in medical engineering requires specific methods of shape analysis and representation that need to be learnt. In this paper the method of knowledge generation as a part of a shape understanding method is proposed. The knowledge generation method used in the system of shape understanding is related to hierarchically organised knowledge of the shape classes. The system of shape understanding that is able to perform different tasks of shape analysis and recognition, based on the ability of the system to understand the different concepts of shape at the different levels of cognition, is proposed. The system consists of different types of experts that perform different processing and reasoning tasks and is designed to perform the visual diagnosis in medical applications. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Mammogram JPEG quantisation matrix optimisation for PACS

    Publication Year: 2001 , Page(s): 1 - 6
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (528 KB) |  | HTML iconHTML  

    Clinical procedures and legal requirements increasingly demand greater performance in JPEG compression of digitised grayscale medical images without loss of visual fidelity or without exceeding a bounded error metric. Current JPEG common practice uses a default general-purpose luminance quantisation matrix. As an initial investigation into defining more suitable quantisation matrices for different medical image modalities, a new candidate matrix has been derived for mammograms. For a given SNR, the new quantisation matrix achieves a relative compression ratio performance improvement of approximately 22% for a quality factor of 95% and 15% for a quality factor of 85%. This study paves the way for a computationally intelligent approach to optimising the quantisation matrix for each medical image modality in both batch mode and adaptively on an image-by-image basis. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A portable device for optically recognizing Braille. I. Hardware development

    Publication Year: 2001 , Page(s): 129 - 134
    Cited by:  Papers (2)
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (609 KB) |  | HTML iconHTML  

    This is the first of a pair of papers that describe a prototype portable device for optically scanning embossed Braille and conversion of the scanned text to binary Braille representation. This prototype has been developed in conjunction with the Association for the Blind (WA). An application to convert the literary Braille code to expanded text has also been implemented and is described in part II. The system developed utilises a hand held scanner that captures the embossed Braille image, in real time, via a linear 128-pixel CCD array. A Texas Instruments digital signal processor performs recognition processing. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Segmentation of clinical structures for radiotherapy treatment planning: a comparison of two morphological approaches

    Publication Year: 2001 , Page(s): 15 - 18
    Cited by:  Papers (3)
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (275 KB) |  | HTML iconHTML  

    The segmentation of therapy-relevant structures is an important but time consuming stage in radiotherapy treatment planning. This paper compares the accuracy of two automatic approaches founded in mathematical morphology by examining their performance in segmenting the bladder in CT studies of the human pelvis. Both automatic approaches are evaluated with respect to a gold standard average manual segmentation derived from delineations produced by five clinical experts. The initial results show that although the segmentations produced by both morphological operators are similar there is still a need for manual interaction. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An algorithm for reducing the effect of compression/decompression techniques on fmgerprint minutiae

    Publication Year: 2001 , Page(s): 243 - 246
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (333 KB) |  | HTML iconHTML  

    We show in this paper that the wavelet transform is an effective technique for significantly compressing fingerprint images, while retaining the necessary structural details for accurate identification. The paper presents a new algorithm for extracting the fundamental identification features for use in automated fingerprint matching systems. Contrary to existing algorithms, the method presented here is tested on compressed/decompressed fingerprint images. We show that using the compression algorithm proposed earlier by the author, we are able to obtain a valid description of the fingerprints at compression ratios down to 50:1. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Teaching undergraduates medical technology innovation and business planning

    Publication Year: 2001 , Page(s): 373 - 378
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (474 KB) |  | HTML iconHTML  

    This paper describes the curriculum and approach used to introduce undergraduate biomedical engineering students to the process of medical technology innovation and business planning. Acknowledging that undergraduates may lack experience in relation to many of the issues that need to be dealt with in relation to the design, development and commercialisation of new technology, the author has progressively used the Internet as a growing window to the world to enable students to build a knowledge base beyond their own experience. Like real first time innovators the students are expected to acquire the knowledge required to firstly generate new design concepts, then select the most medically and commercially interesting design concept, establish IP management requirements, look at costs associated with R&D, product development and commercialisation, identify regulatory requirements, assess competition, and then analyse cost and profit. The classroom teaching is based on recent Australian technology case studies and the innovation focus is concentrated in a prescribed area of medical technology. The desired outcome is a knowledge of the issues that either assist or inhibit the development of new medical technology and insights into the innovation process. Each undergraduate student develops a business plan for the development of a new medical device. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A fast algorithm for image restoration using a recurrent neural network with bound-constrained quadratic optimization

    Publication Year: 2001 , Page(s): 111 - 115
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (382 KB) |  | HTML iconHTML  

    This paper presents a fast algorithm for a recurrent neural network that can restore a degraded image with fewer iterations and shorter processing time by using bound-constrained quadratic optimization (BCQO) and a weighted mask. The BCQO technique has already been used in signal restoration, however implementation of this method in image restoration requires considerable memory and it is computationally expensive. The proposed algorithm replaces the weight matrix of the network with a much smaller mask, thus reducing the processing time and requiring much less memory space. This algorithm produces better results than those obtained by Wiener filter, and achieves image restoration with less iterations compared to a modified Hopfield neural network. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A novel design of integrated proximity sensors for the white cane

    Publication Year: 2001 , Page(s): 197 - 201
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (437 KB) |  | HTML iconHTML  

    The paper documents a low cost solution for integrating proximity sensors onto the white cane for the blind. It promotes the sensors as a peripheral device that uses digital signal processing and ultrasonics to compute distance measurements using time-delay calculations of ultrasonic signals. The design increases the coverage of the cane and overcomes the common bad habits in techniques of its use. The integrated cane will allow detection of high-level objects and give a greater reaction time for changes in the level of the user's path, such as dips. Initial calculations for the timing of signals, and the coverage of sensors for the cane are explored, and descriptions of signal filtering and audio feedback methods used are discussed. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Australian currency note identifier for the vision impaired: Part I hardware description

    Publication Year: 2001 , Page(s): 135 - 139
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (821 KB) |  | HTML iconHTML  

    Polymer notes recently replaced Australia's decimal paper currency. The denominations of the new notes, while very distinctive in their colour and design, are very difficult to differentiate for a vision impaired person. All notes are of equal width and increase only slightly in length with increasing value. The article describes the design and development of a prototype device to identify Australian polymer banknotes, which indicates the denomination to a vision impaired person using a digitally recorded voice output. Use is made of a charge-coupled device (CCD) linear array and a digital signal processing (DSP) chip. Identification is performed by imaging and recognising the contents of the clear window found near the lower corner of each note which is unique for each denomination. This development is of significance in Australia to people who suffer a large degree of vision impairment, and possibly also to the vision impaired population of the European Union, which may also plan to adopt polymer note technology. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Adaptive expanding B-snake model for extracting ultrasound breast lump boundary

    Publication Year: 2001 , Page(s): 19 - 23
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (370 KB) |  | HTML iconHTML  

    In this paper we introduce a new adaptive expanding B-snake model (AEBS) to extract 2-D ultrasound breast lump boundaries, using B-spline to split the whole snake energy into smaller parts to achieve both global deformation and local variation. Due to the noisy property of ultrasound breast lump images, initial model is loaded inside lump area and an expanding force is used to reduce the 'noise' from the breast tissues. Which has been improved after a pre-processing before the AEBS. Local elements' gradient is used as a feedback of contour deformation to make the expanding force to act intelligently. The balance of all defined forces in t-n (tangent and normal) coordinates is used as a terminating criterion. A serial of parallel-scanned ultrasound breast lump images have been used in the testing and results show our method works correctly and consistently. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Australian currency note identifier for the vision impaired: Part II software description

    Publication Year: 2001 , Page(s): 147 - 152
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (544 KB) |  | HTML iconHTML  

    For pt.I see ibid., p.135-40 (2001). The article describes the software component of a prototype currency identifier. Identification is performed by imaging and recognising the contents of the clear window found near the lower corner of each note (I. Siewart, 1998), which is unique for each denomination. This development is of significance in Australia to people who suffer a large degree of vision impairment, and possibly also to the vision impaired population of the European Union, which may also plan to adopt polymer note technology. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Adaptive complex modified probabilistic neural network in digital channel equalization

    Publication Year: 2001 , Page(s): 247 - 251
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (366 KB) |  | HTML iconHTML  

    A novel adaptive technique is proposed for the complex-valued modified probabilistic neural network (MPNN). The adaptive feature is desirable when using the MPNN in channel equalization to track time-varying channels. The MPNN is initially trained using the clustering technique. When training is completed, the network is switched to decision-directed mode and the network parameters are adapted using stochastic gradient-based algorithms in an unsupervised manner. Simulations show that the equalizer was able to efficiently equalize 4-QAM symbol sequences transmitted through nonlinear, slowly time-varying channels. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Level identification using input data mining for hierarchical fuzzy system

    Publication Year: 2001 , Page(s): 379 - 383
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (353 KB) |  | HTML iconHTML  

    Fuzzy rule based systems have been very popular in many control applications. However, when fuzzy control systems are used in real problems, many rules may be required. A hierarchical fuzzy system that partitions a problem for more efficient computation may be the answer. When creating a hierarchical fuzzy system, the level identification stage is crucial and time-consuming. This has a direct effect on how efficient the hierarchical fuzzy system is. This paper reports the use of an input data mining technique to efficiently perform the level identification stage. Without the use of input data mining, k*(k-1) ways of building the hierarchical fuzzy system must be tried. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A study of the shape change of sheep chondrocytes with application of compression to cartilage

    Publication Year: 2001 , Page(s): 95 - 99
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (334 KB) |  | HTML iconHTML  

    In the clinical field, osteoarthritis and most other forms of arthritis are associated with the functional breakdown of articular cartilage. Studies have shown that the mechanical forces exerted on articular cartilage can change its structure and composition, which results in a change in the biomechanical behaviour of the cartilage. Articular cartilage has a unique ultra-structure which provides it with exceptional biomechanical properties. Before the application of laser scanning confocal microscopy (LSCM), studies of the internal structure and mechanical behaviour of articular cartilage had to be carried out independently. Articular cartilage is a load-bearing tissue. To gain real understanding of its biomechanical behaviour, the tissue must be studied in relation to its biomechanical function. Based on the use of a fibre-optic laser scanning confocal microscope (FLSCM), which is slightly different from a traditional LSCM, a methodology has been developed to study the morphology of sheep chondrocytes while the articular cartilage is being compressed. With the characteristics of the FLSCM involved in this study, there is a potential to develop a clinical method for the diagnosis of early internal degeneration of articular cartilage. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A full explanation facility for a MLP network that classifies low-back-pain patients

    Publication Year: 2001 , Page(s): 47 - 52
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (403 KB) |  | HTML iconHTML  

    This paper presents a full explanation facility that has been developed for a standard MLP network, with binary input neurons that performs a classification task. It is shown how an explanation for any input case is represented by a non-linear ranked data relationship of key inputs, in both text and graphical forms. Using the facility, the knowledge that the MLP has learned can be represented by average ranked class profiles or as a set of rules induced from all training cases. The full explanation facility discovers the MLP knowledge bounds by finding the hidden layer decision regions containing correctly classified training examples. Novel inputs are detected by the explanation facility, on an input case-by-case basis, when the case is positioned in a decision region outside the knowledge bounds. Results using the facility are presented for a real-world MLP network that classifies low-back-pain patients. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Speech perception based algorithm for the separation of overlapping speech signal

    Publication Year: 2001 , Page(s): 341 - 344
    Cited by:  Papers (1)  |  Patents (1)
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (320 KB) |  | HTML iconHTML  

    An algorithm for the analysis of speech utilising the time frequency properties of wavelets is introduced. The extracted wavelet coefficients are analysed using two techniques, firstly a covariance matrix is generated to provide information about speaker characteristics. Second, the kurtosis of the wavelet coefficients is used to facilitate the detection of multiple speakers. Preliminary results show that some phonetic information, such as articulation placement and identification of voiced/unvoiced sections, can be extracted from the kurtosis analysis. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Gait symmetry quantification during treadmill walking

    Publication Year: 2001 , Page(s): 203 - 206
    Cited by:  Papers (7)
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (360 KB) |  | HTML iconHTML  

    Normal gait has often been assumed to be symmetric, though several papers have reported the differences in gait parameters between the limbs. A few papers have discussed the temporo-spatial and kinematic symmetry in healthy adults using different statistical and mathematical calculations and no consensus exists on the methodology. The aim of this study was to define symmetry in long term treadmill walking and to compare mathematical symmetry indices. Fifteen minutes of walking of three healthy young adults was recorded bilaterally using two 50 Hz video cameras and temporal gait parameters (stance, swing and stride times) were defined using footswitches. Symmetry between the limbs was analysed using three different indices. The results indicated that despite the symmetry on the stride times, asymmetries were seen in the parameters reflecting within limb phase ratios. The three indices provided very little differences in outcome. When defining symmetry one should carefully select the analysed gait parameters. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Classification of electromyograph for localised muscle fatigue using neural networks

    Publication Year: 2001 , Page(s): 271 - 275
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (416 KB) |  | HTML iconHTML  

    To determine the status of a muscle, surface electromyography (SEMG) is a useful tool being non-invasive and easy to record. Clinicians are able to classify the signal visually but because of the large number of parameters of the signal, automatic classification becomes difficult. This paper reports our efforts at using Wavelet Transforms to process the signal before using Neural Networks for classification. The paper reports that by using specific wavelets for transform and at specific levels of decomposition, the features of the signal correlating with muscle status were highlighted and classification of this data using neural networks gave excellent results. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An electronic design of a low cost Braille typewriter

    Publication Year: 2001 , Page(s): 153 - 157
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (350 KB) |  | HTML iconHTML  

    The paper documents a new design for a Braille typewriter. Comprising a majority of electrical components; the design aims to produce a product that fills the gap in the range of Braillers available. A low cost and robust design will provide the blind with an affordable and reliable alternative to the Perkins and Mountbatten Braillers. The Braille typewriter will comprise the following key components: 10 digit keypad; PIC 16F877 micro controller; 3 stepper motors; 3 stepper motor controllers; embossing tool; 2 solenoids; warning speaker; and nylon casing. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Minerva scene analysis benchmark

    Publication Year: 2001 , Page(s): 231 - 235
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (882 KB) |  | HTML iconHTML  

    The analysis of natural scenes is an important research area. Scene analysis research provides the foundation for the development of autonomous systems whose vision sensors provide important information about the surrounding environment. In this paper we introduce the Minerva scene analysis benchmark to the vision community and provide preliminary results on this data. The scene analysis benchmark contains 448 natural images in both colour and greyscale format. The images contain 8 natural objects including sky, brick, clouds, pebbles, road, trees, grass and leaves. The benchmark is intended to facilitate further research into scene analysis and to encourage the development of tools and techniques that work on natural object recognition. The results reported here have used four image segmentation techniques including fuzzy c-means clustering, histogram based thresholding, region growing, and split and merge. Following segmentation, texture features for object classification have been generated using five different texture analysis methods including autocorrelation, co-occurrence matrices, edge frequency, Law's, and run length. These results can be taken as a preliminary baseline on this benchmark. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Rule extraction based on rough set theory combined with genetic programming and its application to medical data analysis

    Publication Year: 2001 , Page(s): 385 - 390
    Cited by:  Papers (4)
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (553 KB) |  | HTML iconHTML  

    A methodology for using rough sets for preference modeling in decision problems is presented in this paper, where we introduce a new approach for deriving knowledge rules from medical databases based on rough sets combined with genetic programming. Genetic programming is one of the newest techniques in applications of artificial intelligence. Rough set theory (Z. Pawluk, 1982), is nowadays rapidly developing branch of artificial intelligence and soft computing. At first glance, the two methodologies have nothing in common. Rough sets construct the representation of knowledge in terms of attributes, semantic decision rules, etc. On the other hand, genetic programming attempts to automatically create computer programs from a high-level statement of the problem requirements. However, in spite of these differences, it is interesting to try to incorporate both approaches into a combined system. The challenge is to get as much as possible from this association. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Cancer prognosis methodology for cervical cancer to be used by Peter MacCallum Cancer Institute

    Publication Year: 2001 , Page(s): 25 - 30
    Cited by:  Papers (1)
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (986 KB) |  | HTML iconHTML  

    This paper gives the design of techniques for the diagnosis and monitoring of cervical cancer at Peter MacCallum Cancer Institute (Melbourne). These techniques will require an advanced software platform to store and retrieve images from all hospital computers located at different locations. The research will investigate the efficient way of communication between all computers to exchange information using Local Area Network (LAN). Advanced algorithms for analysing stored images are required to help the detection of the degree of the spread of cancer with patients. This project aims at: designing and testing such techniques. The outcome of this research will be a software: system that will be used by doctors of Peter MacCallum Cancer Institute to improve the current cancer detection techniques. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Harmonic distortion at the electrode/electrolyte interface: A physical interpretation of Schwan's empirical model

    Publication Year: 2001 , Page(s): 277 - 280
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (298 KB) |  | HTML iconHTML  

    This paper concerns the modelling and interpretation of the non-linear harmonic behaviour of biomedical electrode/electrolyte interfaces. For the first time a physical interpretation of the published work on this phenomenon by Schwan and his colleagues (1999) is given. Higher order harmonics were calculated based on the assumption that the non-linearity of the interfacial impedance is dominated at low frequencies by the faradaic charge transfer resistance which shunts the relatively linear non-faradaic double layer capacitance. The dependence of harmonic components on the applied signal amplitude was compared between (i) a model based on the Butler-Volmer equation and (ii) Schwan's experimental and empirical results. A good agreement was found, between the theoretical model and Schwan's empirical observations. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.