Prior Expertise Research Articles

Knowledge-based resource management for multifunction radar: a look at scheduling and task prioritization

In this article, we consider two related aspects of radar resource management, scheduling and task prioritization. Two different methods of scheduling are examined and compared and their differences and similarities highlighted. The comparison suggests that prioritization of tasks plays a dominant role in determining performance. A prioritization scheme based on fuzzy logic is subsequently contrasted and compared with a hard logic approach as a basis for task prioritization. The setting of priorities is shown to be critically dependent on prior expert knowledge. By assessing the priorities of targets and sectors of surveillance according to a set of rules it is attempted to imitate the human decision-making process such that the resource manager can distribute the radar resources in a more effective way. Results suggest that the fuzzy approach is a valid means of evaluating the relative importance of the radar tasks; the resulting priorities have been adapted by the fuzzy logic prioritization method, according to how the radar system perceived the surrounding environment.

Fundamentals of Robotics: Linking Perception to Action (Series in Machine Perception and Artificial Intelligence)

9R5. Fundamentals of Robotics: Linking Perception to Action (Series in Machine Perception and Artificial Intelligence). - Edited by Ming Xie (Singapore–MIT Alliance, Singapore). World Science Publications, Singapore. 2003. 692 pp. ISBN 981-238-313-1.Reviewed by RL Huston (Dept of Mech, Indust, and Nucl Eng, Univ of Cincinnati, PO Box 210072, Cincinnati OH 45221-0072).This book is probably best described as a contemporary reference describing the state of the art of robotics, at this time. It is a truly impressive volume where the author documents the many significant robotic advances during the last quarter century. He also looks to the future. The book discusses the broad range of robotics technology ranging from mechanical considerations, to electrical and electronics devices, to control, to vision, to decision making and artificial intelligence. The book is intended for students, for those just getting started in the field, and for experienced practitioners. The book has three underlying themes: (1) a presentation of the fundamental underlying physics and associated analytical procedures; (2) thinking of robotics from a systems perspective; and (3) treating the entire subject tutorially. Throughout the book the author presents the physical principles, the mathematical analysis, applications, and illustrative examples. In many instances he provides extensive explanations of particular aspects of the technology, including design considerations. At all times, the focus is upon viewing a robot as a complex system incorporating a wide variety of technologies, but principally electro-mechanical-control and decision making. The book itself is divided into nine chapters spanning approximately 700 pages. The first chapter provides a brief introduction to the subject outlining the various applications of robots ranging from the industrial/manufacturing setting to the household environment. Contained therein is an interesting discussion about humanoid robots. The author also summarizes the principal issues and problems in robotics technology. The second and third chapters are devoted to mechanical issues such as kinematics (position displacement, rotation, and velocity), mechanisms, joints, and chains. Both forward and inverse kinematics are discussed. The fourth chapter then considers robot dynamics, motors, and drive devices. Control issues are studied in the fifth chapter. The ensuing topics include automatic feedback systems, control elements, sensing elements, design consideration, algorithms, and joint-, task-, and image-space control. In the next three chapters the author discusses information systems, visual sensory systems, and visual perception systems of robots. The ninth chapter is then devoted to decision making including issues of planning, mapping, constraints, and control. The book concludes with a very brief chapter on future expectations. Each chapter contains examples, exercises, and a bibliography. The author is to be commended for the ambitious undertaking of trying to incorporate all these topics into a single volume. The finished product is impressive. The writing is good and the examples are clear and informative. There are, of course, many places where prior knowledge and expertise are needed to fully benefit from the discussion. On balance, however, the book should be of interest and use to those either working in robotics, hoping to enter the field, or those simply having a curiosity about the subject matter. Purchase is recommended both for individuals and libraries.

A Bayesian Network Classification Methodology for Gene Expression Data

We present new techniques for the application of a Bayesian network learning framework to the problem of classifying gene expression data. The focus on classification permits us to develop techniques that address in several ways the complexities of learning Bayesian nets. Our classification model reduces the Bayesian network learning problem to the problem of learning multiple subnetworks, each consisting of a class label node and its set of parent genes. We argue that this classification model is more appropriate for the gene expression domain than are other structurally similar Bayesian network classification models, such as Naive Bayes and Tree Augmented Naive Bayes (TAN), because our model is consistent with prior domain experience suggesting that a relatively small number of genes, taken in different combinations, is required to predict most clinical classes of interest. Within this framework, we consider two different approaches to identifying parent sets which are supported by the gene expression observations and any other currently available evidence. One approach employs a simple greedy algorithm to search the universe of all genes; the second approach develops and applies a gene selection algorithm whose results are incorporated as a prior to enable an exhaustive search for parent sets over a restricted universe of genes. Two other significant contributions are the construction of classifiers from multiple, competing Bayesian network hypotheses and algorithmic methods for normalizing and binning gene expression data in the absence of prior expert knowledge. Our classifiers are developed under a cross validation regimen and then validated on corresponding out-of-sample test sets. The classifiers attain a classification rate in excess of 90% on out-of-sample test sets for two publicly available datasets. We present an extensive compilation of results reported in the literature for other classification methods run against these same two datasets. Our results are comparable to, or better than, any we have found reported for these two sets, when a train-test protocol as stringent as ours is followed.

Fundamentals of Robotics: Linking Perception to Action. Series in Machine Perception and Artificial Intelligence

7R4. Fundamentals of Robotics: Linking Perception to Action. Series in Machine Perception and Artificial Intelligence. Edited by Ming Xie (Singapore–MIT Alliance, Singapore). World Sci Publ, Singapore. 2003. 692 pp. ISBN 981-238-313-1.Reviewed by RL Huston (Dept of Mech, Indust, and Nucl Eng, Univ of Cincinnati, PO Box 210072, Cincinnati OH 45221-0072).This book is probably best described as a contemporary reference describing the state-of-the-art of robotics, at this time. It is a truly impressive volume where the author documents the many significant robotic advances during the last quarter century and also looks to the future. The book discusses the broad range of robotics technology ranging from mechanical considerations, to electrical and electronics devices, to control, to vision, to decision-making and artificial intelligence. The book is intended for students, for those just getting started in the field, and for experienced practitioners. The book has three underlying themes: 1) A presentation of the fundamental underlying physics and associated analytical procedures; 2) Thinking of robotics from a systems perspective; and 3) Treating the entire subject tutorially. Throughout the book the author presents the physical principles, the mathematical analysis, applications, and illustrative examples. In many instances he provides extensive explanations of particular aspects of the technology, including design considerations. At all times, the focus is upon viewing a robot as a complex system incorporating a wide variety of technologies, but principally electro-mechanical-control and decision making. The book itself is divided into nine chapters spanning approximately 700 pages. Chapter 1 provides a brief introduction to the subject outlining the various applications of robots ranging from the industrial/manufacturing setting to the household environment. Contained therein is an interesting discussion about humanoid robots. The author also summarizes the principal issues and problems in robotics technology. The second and third chapters are devoted to mechanical issues such as kinematics (position displacement, rotation, and velocity), mechanisms, joints, and chains. Both forward and inverse kinematics is discussed. The fourth chapter then considers robot dynamics, motors, and drive devices. Control issues are studied in Chapter 5. The ensuing topics include automatic feedback systems, control elements, sensing elements, design consideration, algorithms, and joint-, task-, and image-space control. In the next three chapters the author discusses information systems, visual sensory systems, and visual perception systems of robots. Chapter 9 is devoted to decision-making, including issues of planning, mapping, constraints, and control. The book concludes with a very brief chapter on future expectations. Each chapter contains examples, exercises, and a bibliography. The author is to be commended for the ambitious undertaking of trying to incorporate all these topics into a single volume. The finished product is impressive. The writing is good and the examples are clear and informative. There are of course many places where prior knowledge and expertise are needed to fully benefit from the discussion. On balance, however, Fundamentals of Robotics: Linking Perception to Action should be of interest and use to those either working in robotics, hoping to enter the field, or those simply having a curiosity about the subject matter. Purchase is recommended both for individuals and libraries.

The Heuristic Evaluation of Web-Sites Concerning the Evaluators' Expertise and the Appropriate Criteria List

Research for the evaluation of web-sites has already begun, however it is proceeding at a very slow rate. The main reasons for this are, in our opinion, the attempt to adapt existing methodologies to the particularities of the web, the individual s tructure of web-sites and the issue of finding the appropriate evaluators. This study copes exactly with these points and suggests a heuristic approach for the evaluation of web-sites. In our study we tried primarily to train the evaluators in the particularities of the heuristic evalu- ation; in its classic form as well as in its web-adapted form. By doing this we try to answer the core question if we can augment the evaluators' expertise with a kind of training prior to the conduction of the evaluation itself. Next we used web-adapted heuristics, found in relative literature and tried to clarify them to the evaluators as well. Finally the evaluators were involved in a real evaluation of five web sites and they wrote down their comments on appropriately prepared questionnaires. The results from this study confirm firstly two known conclusions, that the method is applicable to the Web and that the prior evaluators' expertise is of great importance. Yet, in addition to these, we concluded that it is possible to augment, under conditions, this expertise in a short way so they have an increased performance during the evaluation as well. Our main conclusion is, however, that the used heuristic list performed inadequately, but we noted the trend of the evaluators following a somewhat similar mode of thinking, thus providing us with the way to adapt these heuristics in a more holistic approach to the web.

A Bayesian approach for estimating values for prevalence and diagnostic test characteristics of porcine cysticercosis

Several diagnostic techniques are used to estimate the prevalence of the zoonotic tapeworm Taenia solium in pigs, but none of these tests are perfect, making interpretation of results difficult. A Bayesian approach was used to estimate values for the prevalence and diagnostic test characteristic of porcine cysticercosis by combining results of four imperfect tests. Village pigs ( N=868), slaughtered in Lusaka (Zambia), were bled, and tongue and routine meat inspected; and serum antibody and parasite antigen concentrations were determined by ELISA. A model, based on a multinomial distribution and including all possible interactions between the individual tests required 31 parameters to be estimated, but actually allowed only 15 parameters (i.e. had 15 degrees of freedom) to be estimated. Therefore, prior expert opinion on specificity and (in)-dependence of the tests was entered in the model, resulting in a reduction of the number of parameters to be estimated. The estimated prevalence of porcine cysticercosis was 0.642 (95% confidence interval 0.54–0.91). The performances of the tests were (sensitivity (se)–specificity (sp)): tongue inspection (se 0.210–sp 1.000), meat inspection (se 0.221–sp 1.000), Ab-ELISA (se 0.358–sp 0.917), Ag-ELISA (se 0.867–sp 0.947). To validate the estimates obtained from the model we performed a second study: 65 randomly purchased Zambian village pigs were bled for serum antibody and antigen determination, their tongue and meat inspected; and in addition, the carcasses were dissected for total cysticercus counts (gold standard). Cysticerci were found in 31 pigs (prevalence 0.477, 95% confidence interval 0.35–0.60), overlapping with the estimated prevalence in the first study. Sensitivity and specificity values obtained for the aforementioned tests in this study were in agreement with those estimated. A Bayesian analysis framework offers the possibility to combine prior opinion with experimental data to more accurately estimate the real prevalence of porcine cysticercosis in the absence of a gold standard.

The structural qualitative method: a promising forecasting tool for developing country markets

The growth rate model of forecasting used so far by business organizations in developing countries can no longer generate satisfactory results owing to maturing markets. Lack of availability of sufficiently long time series data with these organizations limits the use of time series or causal models. While qualitative methods are suitable in such situations, previous studies have suggested systematic use of information to improve the accuracy of forecasts generated from these methods. This paper develops a new model called the structural qualitative method (SQM), which generates forecasts using quantitative and qualitative data. Application of this model by an Indian seed company is also illustrated here. Using this method forecasts at disaggregate levels can also be generated. It is suitable for use by both large and small organizations and does not require any prior expertise. It can also aid decision-making since some variables are endogenous in the model.

Encoding prior experts judgments to improve risk analysis of extreme hydrological events via POT modeling

One of the main decisions to be made in operational hydrology is to estimate design floods for safety purposes. These floods are generally much rare events that have already been systematically recorded and consequently the results of any estimation process are subject to high levels of uncertainty. When adopting the frequentist framework of probability, the so called ‘respect of scientific objectivity’ shall forbid the hydrologists to introduce prior knowledge such as quantified hydrological expertise into the analysis. However, such an expertise can significantly improve the capability of a probabilistic model to extrapolate extreme value events. The Bayesian paradigm offers coherent tools to quantify the prior knowledge of experts. This paper develops an inference procedure for the peak over threshold (POT) model, using semi-conjugate informative priors. Such prior structures are convenient to encode a wide variety of prior expertise. They avoid recourse to Monte Carlo Markov Chain techniques which are presently the standard for Bayesian analyses, but such algorithms may be uneasy to implement. We show that prior expertise can significantly reduce uncertainty on design values. Using the Garonne case study with a sample of systematic data spanning over the period 1913–1977, we point out that: (1) the elicitation approach for subjective prior information can be based on quantities with a definite practical hydrological meaning for the expert; (2) with respect to the usual Poisson–Generalized Pareto model, a semi-conjugate prior offers a flexible structure to assess expert knowledge about extreme behavior of the river flows. In addition, it leads to quasi-analytical formulations; (3) tractable algorithms can be implemented to approximate the prior uncertainty about POT parameters into these semi conjugate distribution forms via simple Monte Carlo simulations and normal approximations; (4) the design value and its credible interval are notably changed when incorporating prior knowledge into the risk analysis.

University Research Funding: More than Supporting the Best to Do the Best

I would like to offer two components that I think Howard Birnbaum underemphasizes in his accurate and useful discussion of problems that have developed with university research funding. These are my personal observations, distilled from decades of trying to play the game.The first concerns peer review of proposals, which many of us, forgetting that peer review can be no better than the peers, are loath to disparage. In my 40-plus years of experience, the process has changed almost beyond recognition. I remember when peer reviewing was essentially free of self-interest on the reviewer’s part: A proposal showed up in the mail, often submitted not in response to some “announced opportunity” stating that research funds were available, but because an individual investigator had an idea he or she wanted to pursue—and one reviewed it in that light, then sent it back.Today, the process seems anything but individual or disinterested. The peers have a level of awareness of each other and networks of connection and organization that surpass anything imaginable in the 1960s. Potential research collaborators formally combine with a deliberate effort to add so many prestigious names to a proposal that a junior reviewer would have to be very brave to do anything but approve it. The proposers are well aware that they will be among each other’s reviewers and that personalities will matter.The peer review process is enhanced by agencies that advocate “critical mass” and “adequacy of facilities” as necessary criteria for a proposal’s viability. The funding needs for some research groups are so large and so continuous that every announced opportunity is followed up with a proposal, whether or not any prior interest or expertise in the specific area was present in the group. Groups propose first and ask questions afterward, often on a very short time scale; many are still looking for new collaborators a week or two before the proposal is due.Formal but minimal commitments, in terms of time allotted, may be spread over many proposals for a single prominent investigator. Principal investigators will often appear on different multi-institutional proposals in competition for the same funding. That type of competition could only be altered by a new and uniform set of civil-service rules and procedures that the present generation of agency managers would not know how to enforce. Rules for dealing with, regulating, and avoiding conflict of interest within government agencies have been historically rather scrupulously enforced; but application of conflict-of-interest regulations in nongovernmental applications for agency programs seems often to be essentially nonexistent or unenforceable. (And if the agencies tried to implement such regulations, they would be said to be behaving “bureaucratically.”)My other topic is the use of now indispensable graduate student labor in university research programs, usually programs supported by federal grants. There is no cheaper source of high-quality labor, and it frequently comes with a level of motivation not available at any price. Graduate students are intensely interested in the research and would like to pursue it after they finish their education. It is insufficient to tell them when they are applying to graduate degree programs that the buyer should beware and that opportunities to continue doing basic research may be in very short supply when they finish their PhDs. Such warnings only filter out the brightest American students, who are typically better attuned to the practical consequences of their educational choices and better able to choose from a wide range of possibilities—the students who should continue doing basic research in physics for the rest of their lives. The productivity of any research group is surely much greater if several graduate students are involved. The only mechanism I can imagine that would end this practice would be to give funding agencies the authority and competence to stop rewarding universities for the overproduction of PhDs in physics. It would not be an easy task, and the overall national research effort would go more slowly. But difficult problems sometimes do not have easy solutions.© 2002 American Institute of Physics.

Bayeswatch: an overview of Bayesian statistics.

Increasingly, clinical research is evaluated on the quality of its statistical analysis. Traditionally, statistical analyses in clinical research have been carried out from a 'frequentist' perspective. The presence of an alternative paradigm - the Bayesian paradigm - has been relatively unknown in clinical research until recently. There is currently a growing interest in the use of Bayesian statistics in health care research. This is due both to a growing realization of the limitations of frequentist methods and to the ability of Bayesian methods explicitly to incorporate prior expert knowledge and belief into the analyses. This is in contrast to frequentist methods, where prior experience and beliefs tend to be incorporated into the analyses in an ad hoc fashion. This paper outlines the frequentist and Bayesian paradigms. Acute myocardial infarction mortality data are then analysed from both a Bayesian and a frequentist perspective. In some analyses, the two methods are seen to produce comparable results; in others, they produce different results. It is noted that in this example, there are clinically relevant questions that are more easily addressed from a Bayesian perspective. Finally, areas in clinical research where Bayesian ideas are increasingly common are highlighted.

Protein lipid overlay assay.

The Protein Lipid Overlay (PLO) assay enables the identification of the lipid ligands with which lipid binding proteins interact. This assay also provides qualitative information on the relative affinity with which a protein binds to a lipid. In the PLO assay, serial dilutions of different lipids are spotted onto a nitrocellulose membrane to which they attach. These membranes are then incubated with a lipid binding protein possessing an epitope tag. The membranes are washed and the protein, still bound to the membrane by virtue of its interaction with lipid(s), is detected by immunoblotting with an antibody recognizing the epitope tag. This procedure requires only a few micrograms of protein and is quicker and cheaper to perform than other methods that have been developed to assess protein-lipid interactions. The reagents required for the PLO assay are readily available from commercial sources and the assay can be performed in any laboratory, even by those with no prior expertise in this area.

Protein Lipid Overlay Assay

The Protein Lipid Overlay (PLO) assay enables the identification of the lipid ligands with which lipid binding proteins interact. This assay also provides qualitative information on the relative affinity with which a protein binds to a lipid. In the PLO assay, serial dilutions of different lipids are spotted onto a nitrocellulose membrane to which they attach. These membranes are then incubated with a lipid binding protein possessing an epitope tag. The membranes are washed and the protein, still bound to the membrane by virtue of its interaction with lipid(s), is detected by immunoblotting with an antibody recognizing the epitope tag. This procedure requires only a few micrograms of protein and is quicker and cheaper to perform than other methods that have been developed to assess protein-lipid interactions. The reagents required for the PLO assay are readily available from commercial sources and the assay can be performed in any laboratory, even by those with no prior expertise in this area.

Expression Profiling of Human Tumors: The End of Surgical Pathology?

Presented by Marc Ladanyi Expression profiling is a new investigative and diagnostic modality made possible by recent scientific and technological advances (for recent reviews, see 1, 2 ). The key scientific advance is the Human Genome Project, in particular, its so-called “expressed sequence tag” (EST) component, which focused on collecting the sequences (in most cases only partial, i.e., “sequence tags”) of all of the genes (both known and unknown) expressed in a wide variety of human tissues. By now, the majority of named and anonymous human genes are represented respectively in the GenBank and the EST sequence databases. The key technological advance has been the miniaturization and automation of molecular biology, specifically the advent of DNA microarray technology. DNA microarrays can be made on either nylon membranes, glass slides, or synthetic “chips,” and can consist either of cDNA clones or cDNA clone-specific oligonucleotides. The numbers of DNA spots on microarrays can number from the hundreds to the tens of thousands. Expression profiling refers to the process of establishing the pattern of expression of thousands of individual genes simultaneously in a given cell or tissue sample by extracting its RNA, converting it to cDNA and hybridizing the labeled cDNA to a DNA microarray. In a sense, pathologists already use gene expression studies diagnostically on a daily basis in the form of immunohistochemistry. Expression profiling provides a supraexponential increase to the amount of gene expression data available on a given tumor, but lacks the topographical information of immunohistochemistry. Nonetheless, it is widely believed that the latter shortcoming will be more than compensated by the sheer multiplicity of the gene expression data. It is already apparent that expression profiling can be used to distinguish different human tumors. With fascinating exceptions, 3 some of the differential diagnoses achieved so far by this approach have been rather straigthforward by the standards of histopathology. 4, 5, 6, 7, 8 Thesetypes of microarray-based “assays” have not yet been subjected to a rigorous evaluation of sensitivity, specifity, and cost-effectiveness as a diagnostic modality. However, ongoing developments such as the rapid and continual scientific refinement of expression profiling, the possibility of mass production of smaller more narrowly targeted microarrays, and the potential for automated software-based recognition of specific gene expression patterns, promise to address many of these concerns and present limitations. Aside from its diagnostic applications, expression profiling can also provide important prognostic information. 6 It is evident that morphology alone, while quite useful for the diagnostic classification of tumors, does not reflect the entire biology of a given tumor. Tumors of similar appearance can behave differently and expression profiling is expected to predict tumor behavior and response to chemotherapy more reliably than tumor histopathology. In recent years, the impact of diagnostic molecular genetic markers, typically based on the detection of specific translocations, has been felt most acutely in hematopathology and sarcoma pathology, where the concept of morphological/genetic entities has become well accepted. Expression profiling promises to extend this process to all tumors, including carcinomas. A key difference is that, as a more “global” approach, expression profiling may require less prior expert histopathological “triage” that any prior ancillary diagnostic technique. Will expression profiling evolve into another “ancillary” technique or will it come to be seen as an alternative to expert surgical pathology? At present, expression profiling studies are highly dependent on histopathology for validation, 1, 9 but it is conceivable that expression profiles may eventually become “disconnected” from the underlying morphology, assuming the role of clinically relevant entities. The extent of involvement of the pathology community in the further development of microarray-based expression profiling as a diagnostic test may determine the nature of its ultimate relationship to conventional tissue-based diagnosis. 9

Heuristic approach for assigning workers to tasks based on individual learning rates

A heuristic worker-task assignment based on individual worker learning rates is examined for two tasks, one with a long production run, the other with a short production run. Simulations of total productivity are performed under several experimental conditions based on empirical industrial worker productivity measures. Results indicate that the heuristic method significantly improves overall productivity under empirically observed conditions and under many experimental conditions. The heuristic policy performs best in conditions where workers learn more gradually. Conditions are discussed where the heuristic policy provides the greatest potential for improvement based on factors including the mixture of production cycles, mean learning rate, mean forgetting rate, mean prior expertise, variance of prior expertise and variance of steady-state productivity.

Modelling the effects of environmental conditions on apparent photosynthesis of Stipa bromoides by machine learning tools

Apparent leaf photosynthesis of the grass Stipa bromoides was measured in the field in two sites of Northern Greece. For predicting apparent photosynthesis from irradiance, temperature and relative air humidity data, we applied and compared two modelling approaches: ordinary statistical modelling and automatic model construction based on machine learning procedures. Ordinary statistical models were constructed based on background knowledge concerning the response of photosynthesis to irradiance. A Michaelis–Menten type light saturation curve was selected among six candidate models and was extended to include air temperature effects. A bell-shaped function of temperature was substituted for the parameter describing the asymptotic maximum photosynthesis. The final model accounted for 67.3% of data variation and was further improved by splitting the data set by experimental site. Site-specific differences were detected regarding the half saturation constant for light and the optimal temperature for photosynthesis. Automatic model construction produced a number of regression trees that enabled a detailed but simple description of the way irradiance, temperature and relative humidity affect photosynthesis. Photosynthesis increases with increasing irradiance, temperature affects photosynthesis when irradiance is close to saturation levels and relative humidity has an effect when both irradiance and temperature are high. There is a threshold value of relative humidity (at about 35%), below which photosynthesis is independent of irradiance within the observed range, decreasing with increasing temperature when temperature is high (>25°C) and increasing with increasing relative humidity when temperature is low (<25°C). The machine learning tools we used provide a very powerful modelling alternative to ordinary curve fitting methods. Their major advantages are the flexibility to select between accuracy and generality and their robustness against outliers and mixtures of differential responses. The models are transparent and easily interpreted. They seem to be able to handle quite complex dependencies among attributes, not requiring prior expert knowledge.

Effect of a word prediction feature on user performance

his study examines the influence of a word prediction feature on several measures of user performance during text entry tasks. Fourteen subjects transcribed text with and without a word prediction feature for seven test sessions. Eight subjects were able-bodied and used mouthstick typing, while six subjects had high-level spinal cord injuries and used their usual method of keyboard access. Use of word prediction significantly decreased text generation rate for the spinal cord injured (SCI) subjects and only modestly enhanced it for the able-bodied (AB) subjects. While fewer selections were required with the word prediction system, each selection took significantly longer to make, which we attribute to the additional cognitive and perceptual processes associated with use of word prediction. Performance was further analyzed by deriving subjects' times for keypress and list search actions during word prediction use. All subjects had slower keypress times during word prediction use as compared to letters-only typing, and SCI subjects had much slower list search times than AB subjects. Under the conditions of this experiment, the cognitive cost of using word prediction largely overwhelmed the benefit provided by keystroke savings. SCI subjects appeared to incur higher cognitive costs than AB subjects, possibly due to their prior expertise in letters-only typing.

170 Potential time savings using computer assisted segmentation of pelvic CT images

The aim of this study was to evaluate possible savings in the time spent segmenting pelvic CT data for 3 D conformal prostate radiotherapy planning. Three male subjects with early carcinoma of the prostate underwent an extended CT scan of the pelvis. Contiguous 5mm. axial scans (mean 76 images per subject) were acquired from L3 to mid thigh. These images were segmented manually into prostate, bladder, rectum, femoral heads, external contours and pelvic bones. The task was repeated using computer segmentation algorithms developed within the COVIRA project of the EU. These include region detectors, edge detectors, algorithms with both functions and others capable of refining approximate outlines. The planner had prior expert knowledge of these and the most appropriate one for a given tissue. Times reported are the average of 3 segmentations for each of the 3 subjects and include editing: BladderProstateRectumFemoralPelvicContoursHeadsbonesManual5m 20s5m 38s5m 37s9m 5s52m 36s47m 50sComputer2m 40s6m 1s6m 52s6m 14s28s1m 12s Computer segmentation algorithms can considerably reduce the time taken to define pelvic anatomy relevant to radiotherapy treatment planning, thereby increasing the productivity of the planner. They are best applied to those structures which have strong edges and are less well suited to definition of the prostate (poor edge contrast) and rectum (gas filled lumen). More sophisticated algorithms are required if further time savings are to be realised.

Tuning of fuzzy models by fuzzy neural networks

It is relatively easy to construct a rough fuzzy model with expert knowledge. It is difficult, however, to fine-tune the parameters of the fuzzy model in order to get improved behavior. For the purpose of tackling this problem, we propose a fuzzy neural network model. The proposed model utilizes a prior expert knowledge for target systems, and embodies fuzzy models which consist of fuzzy rules whose antecedent and consequent are fuzzy sets. The model is equipped with a fuzzy inferencing and tuning mechanism for model parameters by learning. It allows us to tune such parameters of fuzzy models as linguistic terms and relative rule importance. In addition, to show its applicability, we perform some experiments and present the results.

Lessons in developing mathematical courseware

THE RENAISSANCE PROJECT, funded by Apple Computer, set out to demonstrate the potential of interactive multimedia for University teaching. In an attempt to exploit possible inter-disciplinary requirements, the Renaissance Teams were deliberately chosen to foster cross curricular approaches. The aim of the project was to investigate the use of hyper/multimedia in creating a learning environment which was beneficial to both teachers and learners. This paper describes the experiences of the Mathematics Team, beginning with the development of a hypertext version of an existing textbook and leading to the writing of seven modules at the basic calculus level. Various lessons and principles for courseware writing were learned, which have since been applied in subsequent projects to help academics without prior software expertise to become better courseware authors.

A PC database to facilitate treatment of pediatric hematology/oncology patients.

In an effort to simplify chemotherapy dose calculations and reduce the incidence of medication errors, a commercially available, general database management program was used to aid in the management of an active pediatric hematology/oncology service. The software was customized to include all medications used by the specialty practice, as well as some used in a general pediatric setting. After entering a patient's height and weight, the database rapidly computes and immediately displays the patient's surface area, appropriate intravenous fluid rates, and doses of antibiotic, chemotherapeutic, and other drugs. No prior computer expertise is required to run the program, which can be used easily by physicians and nurses. A database program similar to this can be implemented and customized easily for any hematology/oncology practice, and can facilitate patient care by reducing both the time and effort needed to order medications as well as the number of medication errors.