ࡱ> 5@ bjbj22 "XX"C MMM&Nd|O|O|OPFpFpFp8~pqTPbq.t"tXuXu3vxyy`$&RxC|OT3v3vTTC|O|OXuXuXĊĊĊT|OXu|OXuĊTĊĊd|O|O)XuVq )eFpŰ4in00  h)O$P^NO^|O|O |O)@y}ĊDmyyyCCPPFpPPFpTowards a Language of Health Prepared for Enterprise Strategy Veterans Health Administration Department of Veterans Affairs 810 Vermont St., N.W. Washington, DC. 20420 Tom Munnecke Science Applications International Corporation 10260 Campus Point Ct. San Diego, Ca. 92121 (858) 756 4218  HYPERLINK "mailto:munnecket@saic.com" munnecket@saic.com Version 1.1 December 12, 2001 Available at HYPERLINK "http://www.munnecke.com/papers/D21.doc"http://www.munnecke.com/papers/D22.doc Table of Contents  TOC \o "1-3" \h \z \u  HYPERLINK \l "_Toc532805134" Towards a Language of Health  PAGEREF _Toc532805134 \h 1  HYPERLINK \l "_Toc532805135" Summary  PAGEREF _Toc532805135 \h 2  HYPERLINK \l "_Toc532805136" Introduction  PAGEREF _Toc532805136 \h 2  HYPERLINK \l "_Toc532805137" Ways of Understanding  PAGEREF _Toc532805137 \h 2  HYPERLINK \l "_Toc532805138" Normal and Abnormal Views of Medicine  PAGEREF _Toc532805138 \h 2  HYPERLINK \l "_Toc532805139" Our Current System is Based on Understanding Failures  PAGEREF _Toc532805139 \h 3  HYPERLINK \l "_Toc532805140" Genos: A Framework for Health Information  PAGEREF _Toc532805140 \h 3  HYPERLINK \l "_Toc532805141" Historical Perspective - It All Started with Aristotle  PAGEREF _Toc532805141 \h 3  HYPERLINK \l "_Toc532805142" An Example of Aristotelian Thinking  PAGEREF _Toc532805142 \h 4  HYPERLINK \l "_Toc532805143" Newtons Laws of Motion  PAGEREF _Toc532805143 \h 4  HYPERLINK \l "_Toc532805144" Why Use Aristotelian Doctrine in Todays Thinking?  PAGEREF _Toc532805144 \h 4  HYPERLINK \l "_Toc532805145" Spreadsheets: An Example of a Higher Level Framework  PAGEREF _Toc532805145 \h 5  HYPERLINK \l "_Toc532805146" A Framework for Higher Level Abstraction In Health Science  PAGEREF _Toc532805146 \h 5  HYPERLINK \l "_Toc532805147" What Genos Might Look Like  PAGEREF _Toc532805147 \h 5  HYPERLINK \l "_Toc532805148" Desired Traits of Genos  PAGEREF _Toc532805148 \h 6  HYPERLINK \l "_Toc532805149" Dealing with the Void  PAGEREF _Toc532805149 \h 6  HYPERLINK \l "_Toc532805150" Dealing with the Infinite  PAGEREF _Toc532805150 \h 6  HYPERLINK \l "_Toc532805151" Jumping Out of the Model  PAGEREF _Toc532805151 \h 7  HYPERLINK \l "_Toc532805152" Some Properties of Genos  PAGEREF _Toc532805152 \h 7  HYPERLINK \l "_Toc532805153" Where to go from here?  PAGEREF _Toc532805153 \h 8  HYPERLINK \l "_Toc532805154" Conclusion  PAGEREF _Toc532805154 \h 9  Towards a Language of Health I speak Spanish to God, Italian to women, French to men, and German to my horse Charles V of France Summary This paper discusses a high level conceptual framework for health which is based on emerging discoveries in the fields of genomics, complexity theory, and other disciplines. It discusses some of the difficulties involved with defining disease according to Aristotelian hierarchies that are prevalent in todays terminologies and knowledge bases. It proposes an alternative framework based on the generative properties of normal health, within which emerging discoveries of genomic and proteomic insights can be applied to the clinical process. Introduction Language is the essential force that shapes and describes our thoughts and actions, in medicine as well as in our entire culture, both explicitly and implicitly. This paper examines some of the most crucial and complex issues facing the healthcare industry from a perspective of language and explores the feasibility of creating a new linguistic framework to serve as a kind of cognitive bootstrap, giving us more insightful ways to speak of health and genomic information. Ways of Understanding Our current approach to informatics and health care is based in a paradigm that, like medicine itself, focuses on finding and fixing what is wrong rather than building on what is right. This deficit-focused cultural bent can be instructively illustrated by casting an alien eye at how we deal with technology and its ramifications. For example, let us suppose that Martians studying a television set remove a tube and discover that the TV begins making a humming noise. Repeating the experiment with other receivers and other experimenters, they determine that every time they remove this tube, the TV hums, and every time they plug it back in, the TV stops humming. They have discovered an objective, repeatable, refutable scientific fact. Naming this the anti-hum tube, they go on to map other components of television sets in the same way, arriving at a complete map of anti-failure components and the associated disorders their removal triggers. Having accomplished this task, they then naturally look for second-order problems, i.e., what happens when two components fail simultaneously? Are there underlying syndromes of failure that are somehow related? Our Martian tube-explorers develop an ever-growing catalog of failures and associated components that becomes the basis for future research and testing. This catalog of deficits serves as a powerful force for the future of Martian TV-disorder understanding. The findings and patterns of previous research into tube failures guide future researchers as they seek to understand the complexities of the TV set. New problems are seen as variations on previous failure-related problems; discussions between the various specialists begin to fall along party lines focused on each specialtys view of how TVs fail. Eventually, new sublanguages develop, paralleling the subspecialties of the researchers intent on understanding TV technology. At some point, as the Martians grow in their analysis of television failures, the introduction of new knowledge serves only to create more uncertainty about the gaps in their understanding than it does certainty about knowledge of a specific failure. A tipping point has been reached at which information flips from being a source of order and understanding for the Martians to being a source of disorder and ignorance the result of attempting to understand the functioning of a complex system by looking at what is wrong with it. The Martians focus on failure led them into an indefinite downward spiral, perhaps limited only by the proportion of the Martian National Product they wished to spend on their increasingly costly efforts to diagnose tube failure. It is possible to flip to a way of understanding by examining what is right with the TV. For example, if the Martians abducted an Earthling engineer, he or she could reveal in straightforward terms how a TV works. Gaining knowledge by focusing on how TVs work would make it much easier to determine how they fail. Note the asymmetry between these two approaches to understanding. The Martians could easily move from the tackling how televisions work to how they fail. The opposite, however, is not true. Knowing all the failures in excruciating detail does little to advance the understanding of how a system works. Normal and Abnormal Views of Medicine The difference between the two approaches is illustrated in a paper discussing emerging reference models for medical informatics. It describes two types of reference models, normal and abnormal: Abormal emerging reference terminologies include: Medications (GCPR) MMHCC (NCI) Genes, Proteins, Diseases (NCI) Snomed-RT (CAP) Current Procedural Terminology 5 (AMA) Normal emerging reference terminologies include: Biology of life (yet to be developed) Anatomy (U. of Washington Digital Anatomy Project) Virtual Human (Physiology) a current vision. Our goal in this paper is to outline a normal framework for describing aspects of health. Our Current System is Based on Understanding Failures Similar to the case with our Martian television adventure, our health care system is overwhelmingly based on the failure mode of inquiry, in that it focuses on diseases and problems to solve. The above list of normal and abnormal terminologies illustrates this. The abnormal terminologies have been developed with far greater specificity and resources than the normal ones. Diagnosis of disease is the cornerstone of allopathic medicine, with systems defined as a problem to be solved. This leads to a conundrum: If we deal only with problems, deficits, and diseases, how do we ever value positive, constructive, generative activities not rooted in the negative? Like the Martians pulling tubes from TV sets, the healthcare system is lost in a fog of problems and fixes which take so much of its resources and attention that it cannot see the more general and positive aspects inherent in health and growth. This conundrum is particularly acute with respect to our understanding of the human genome. If we treat our investigation of genomics the way the Martians treated their investigation of TV sets, will we ever be able to rise above the clatter of detecting whats wrong? If our understanding is limited to knocking out genes and discovering disorders, will we have any greater success than Martians pulling out tubes and naming problems? Our understanding of the genome by its expressed failures would definitely create a tremendous number of facts, firmly established according to the scientific method. But rather than leading us to greater knowledge of health and living systems, it could lead us into an ever-deeper spiral of fixation on what can go wrong, rather than building on what is right. The lure of this tremendous quantity of information may hide our understanding of what it truly means to be healthy. Unfortunately, it is not possible to abduct an alien to give us a simple normal understanding of health. We must somehow bootstrap our knowledge of health from our current position of relative ignorance, fueled by an intense desire to learn how things work, rather than how they fail. This paper postulates the potential for developing a framework within which an understanding of health and living systems could be expressed. This new framework would be able to capture notions of growth, cascades, scale, and interaction in descriptive terms not possible in todays fragmented and disease-based language of health. It would provide a framework within which genomic information, clinical information, and personal information could coexist and evolve over time. Genos: A Framework for Health Information Our proposed new hypothetical framework, Genos, would be able to express meaning across a broad range of issues relating to health and living systems. Many of this frameworks desired traits would be rooted in the knowledge and technological achievements gained recently in the fields of fractal theory, chaos theory, genomics, complexity theory, linguistics, and computers. Medical informatics scientists often invoke Aristotelian concepts and hierarchies when struggling with the issue of medical nomenclatures, and this is a paradigm that Genos must overcome. Bertrand Russell once observed that since the seventeenth century most intellectual advance has had to overthrow Aristotelian doctrine. Taking that as guidance, this paper presents a model for our health care system that overthrows many of the concepts of Aristotle. One way to envision Genos is to mentally picture a giant spreadsheet. Fields in this spreadsheet are Genos expressions, much like current spreadsheets can have a cell that represents a formula. Unlike current spreadsheet systems, however, Genos simultaneously is able to express much more complicated relationships of growth, cascades of effects, and interaction over time that are not visible at a given snapshot. Users can browse more deeply to find more information and explore information in richer ways than currently expressible in the structure of todays medical information systems. Genomic information may create entirely new ways of looking at the clinical process, wildly different from those we now use. For example, it may not be possible to express complex genomic information on paper. Time and cascade effects may be introduced that are not describable in terms of simple hierarchies, but rather can be expressed as interaction with computer models. Our advances in genomic understanding could cause a continuous churning of medical knowledge, far greater than anything we have experienced in the past. Attempting to stuff this continuous flow of information into our old containers simply may not be feasible. Genos is a hypothetical framework that, in theory, would be able to deal with this explosion in genetic science. Its development would stimulate thought and discussion of our intellectual infrastructure for dealing with health information in the next century an approach that would build on what is right rather than continue the current focus on fixing what is wrong. Genos would allow us to grow in our understanding and our use of this tremendous volume of new information. Historical Perspective - It All Started with Aristotle The field of medical informatics relies on numerous aspects of Aristotelian thinking for the classification of medical nomenclature. Aristotles ideas are invoked as a foundation for expressing clinical data and as a measure of the formality of the UMLS Metathesaurus: Aristotelian classification requires that each term within a type hierarchy be defined by genus (the category of classification for a term) and differentia (the elements, features, or factors that distinguish one term from another), and that syllogisms be used to analyze the properties inherited by each type. To create an Aristotelian hierarchy, differentia for every term must be specified. Description logic is used to specify these differentia by defining relations and selection constraints that are appropriate for each term, but that differentiate these terms from their immediate parents. These differentia are contained within concept definitions, statements that incorporate both the genus and differentia of each term. This Aristotelian approach has led to a plethora of discussions, committees, and standards efforts, creating in turn a vast array of nomenclatures and terms broken into genera and species (perhaps comparable to all those television tubes pulled out by the Martians). However enduring Aristotles ideas and influences over the past 2400 years, his ideas have not been without controversy. Bertrand Russell said, In spite of the genius of Plato and Aristotle, their thought has vices which proved infinitely harmfulSince the beginning of the seventeenth century, almost every serious intellectual advance has had to begin with an attack on some Aristotelian doctrine.  An Example of Aristotelian Thinking It is instructive to examine some of Aristotles thoughts to better understand the doctrine that governs medical informatics today. Aristotles conceptualization of the universe provides a telling example of the rigidity of Aristotelian concepts: The Aristotelian cosmos had the planets moving in crystalline orbs. However, since there is no infinity, there cant be an endless number of spheres; there must be a last one. This outermost sphere was a midnight blue globe encrusted with tiny glowing points of light the stars. There was no such thing as beyond the outermost sphere; the universe was contained in a nutshell, ensconced comfortably within the sphere of fixed stars; the cosmos was finite in extent, and entirely filled with matter. There was no infinite; there was no void. There was no infinity; there was no zero. Motion was a genus to his way of thinking, which was further differentiated into species of circular and straight. These divisions lead to an elaborate set of conclusions: In fact, Aristotle needed only the distinction between the two types of natural motion, circular and straight, to be ready to tell us what it is like to be a cosmos. First, the nature of circular motion proved for him that the cosmos must be finite. Second, this finite universe is divided into two distinct regions, the upper part, or the region of celestial spheres, where the circular motion reigns supreme, and the region interior to the orbit of the moon (sub luminary) filled with ordinary matter whose nature is to move up or down. Third, since motion reveals the nature or substance of things, the celestial spheres and bodies, stars and planets must be composed of a material as different from ordinary matter, as circular motion is from rectilinear. The ether, as this heavenly substance is called, is therefore a material whose nature is to issue in a uniform circular motion. Fourth, it also follows from the Aristotelian analysis of uniform circular motion that the ether is unalterable, suffers neither growth nor diminution, and has no beginning or end, which is to say that it can be neither generated nor corrupted. Fifth, to show that only one such substance can exist in the universe, Aristotle drew on the concept of nature and purpose and declared that a second substance of this type would be as pointless as a shoe is pointless when it is not worn. But God and Nature create nothing that is pointless." Aristotles ideas were very compelling and reigned for centuries. As Karl Popper states, Erroneous beliefs may have an astonishing power to survive, for thousands of years, in defiance of experience Indeed, the Bishop of Paris decreed in 1277 that God cannot move the heavens in a straight line, because that would leave behind a vacuum. As a 2400-year-old theory, this is a quaint and entertaining explanation of the cosmos. Aristotles approach, however, does not bode well as a template for thinking about other problems, such as medical informatics yet it continues to influence medical informatics thinking and nomenclature. We must question this approach in medical informatics, just as Newton questioned Aristotles theories of the universe. Newtons Laws of Motion With the full weight of ancient philosophers and contemporary authorities behind the Aristotelian model of the cosmos, it was difficult for thinkers to jump outside the prevailing model. Yet that is just what some of them did and one in particular remains a potent influence on scientific thought. Isaac Newton was one of those who worked to overthrow Aristotles categories and ways of thinking, because scientific evidence led him elsewhere. Newton, as we all now know so well, developed and extolled the concepts of zero and infinity in his creation of calculus. His mind leapt from the apocryphal falling apple to a body falling around the earth in orbit, to billiard balls. Out of this, he came up with a four-byte expression that changed the world: F=MA. His theories, and many other successes modern science has achieved, came from jumping outside the genus/species/differentia model Aristotle proposed. F=MA was not a sub language to the Aristotelian model. Rather, it was a totally new model in itself. If we place the known advances in our scientific understanding since the time of Aristotle on a 2400-year timeline, we would find the bulk of them clustered at the very last decades and centuries. Consequently, it remains doubtful that basing the future of medical informatics on Aristotelian thinking is a path to success. In fact, as Russell has pointed out, clinging to Aristotelian doctrine, comforting as it may be to some, actually may prove harmful to our progress. Why Use Aristotelian Doctrine in Todays Thinking? Medical informatics literature tends to reflect the Aristotle-like goal of a crystalline orb of medical nomenclature and terminologies, often employing such terms as consistency, closure, and completeness. The Danish Health Care Classification system, for example, seeks to create a single hierarchy that will provide one coherent system with clear interfaces between sub classifications, one concept one code, consistent hierarchical structure, unambiguous and homogeneous codes, clear and pure division of classes, and fast and consistent updating. The idea that there could be medical knowledge that does not fit into these structures is considered to be anathema: The controlled vocabulary should reject the use of not elsewhere classified terms. True to Aristotles doctrine, rather than accepting voids and infinities, medical informaticians often seek to limit discourse to that which is expressible in the fixed scheme. How can medical knowledge grow, if terms must first be encoded in a dictionary? Dictionaries are historical documents, not prospective guides to meaning. Ignoring observations that are inconsistent is one way of achieving consistency, but is not a good way of growing our understanding health care. Spreadsheets: An Example of a Higher Level Framework Let us employ an example to illustrate the rigidity of current medical informatics and the potential benefits of a richer, more flexible framework for understanding health. Before the days of personal computers and spreadsheets, corporate information was collected and disseminated through fixed programs designed by a central data processing department. Since it was expensive to write these programs, it was critical to make sure they presented data in the most easily comprehensible way to the broadest number of users. A committee would be formed to work with systems analysts to define the best way to format the data. The more heavily used the report, the larger (and, typically, contentious) the committee became. The group had to define the One Correct Way to display the data. The introduction of the spreadsheet changed this dramatically. Committees no longer had to decide how individuals would view data. Instead, users could reformat, graph, add columns, hide irrelevant columns, and do their own analysis with simple clicks of the mouse. The spreadsheet obviated the need for organizational paraphernalia, standards, committees, and associated costs. It dissolved problems, rather than solving them. We can look at the spreadsheet as a higher-level framework of the organizations information processing requirements a framework that provides an overarching ability to meet users needs without having to resort to the predefined, prestructured approach of the central data processing department. This framework provided a method to deal with concepts that arch over a multitude of sub-concepts appearing at a lower level to be disconnected. A spreadsheet user can make the same program operate on an expense report, a monthly admissions summary, a departmental roster, or a golf handicap list. While spreadsheets have helped make data more understandable, usable and flexible, the volume and complexity of medical knowledge now, and in the future, necessitate development of a higher-level framework for medical informatics, one that can help us make the most effective and wisest use of the wondrous advances tantalizingly offered by modern medical science, and especially genomics. In the same way in which spreadsheets obviated much of the organizational paraphernalia associated with data processing systems, Genos could help healthcare systems simplify, streamline, and make their systems more adaptive. A Framework for Higher Level Abstraction In Health Science Newtons work can be viewed as a higher level framework for the understanding of the universe. With it, people were provided the ability to talk about apples falling, moons orbiting the earth, and billiard balls on frictionless tables. Newton created a higher level of abstraction for the universe, just as spreadsheets created a higher level of abstraction for data processing applications. Is it possible to create a similar framework that can express a higher level of abstraction in health? Is it possible that the many apparently conflicting factions of the health care industry, public health concerns, and advances in life sciences and technology can be brought together under a new, overarching framework? This is the ambitious goal of Genos. There is no such framework today, and if it were to appear out of the blue it likely would be met with much skepticism. In the same way that data processing standards committees scoffed at the spreadsheet, and Aristotelian cosmologists fought the ideas of Newton, those who have their careers and educations defined by pre-Genos frameworks would not necessarily find Genos to be a graceful improvement. Genos would deal with voids and infinities that the pre-Genos thinkers may never have considered. The concept of Genos conjures up concerns of chaos and uncertainty for rigid thinkers, who would see the pre-Genos world as ordered and probable, with splits between groups in nomenclature and definition being addressed by interface committees. In fact, in some cases, the interface groups likely have become more active than the functional groups. This, the pre-Genos thinkers would say, is an indication that they are addressing the real problems. The pre-Genos thinkers are so busy doing their integration work that they would have little time to invest in such pie-in-the-sky ideas as a more generalized approach not anchored in fixing things. Post-Genos thinkers, in contrast, would use these very arguments in favor of Genos. The exploding costs of integration, they would contend, are a sign of the lack of higher-level conceptualization. Rather than adding more data processing committees to standardize and interface all existing data processing applications, they would point out, an overarching, spreadsheet-type approach would be much more flexible and effective. What Genos Might Look Like One way to imagine Genos is to think of an entire medical information system, (including the terminologies and metadata) as a kind of giant spreadsheet, which we will call an M-Sheet. Spreadsheet cells have a displayed value, but may have an expression underlying them that is actually a formula, such as the sum of a range of cells. Imagine that each field in the M-Sheet is a Genos expression that normally displays as a single value. Users of the medical information system can operate it normally, perhaps not even being aware that Genos is controlling the fields in the background. This is analogous to spreadsheet users who may not be aware of all the spreadsheet formulas working on their data. The M-sheet can be viewed as a surface level expression of a much deeper, more complex set of expressions. Genos may be dealing with growth processes that change dramatically over time and in response to history and environment, yet it may report only the snapshot value for the M-sheets evaluation. For example, an M-sheet field may hold information on whether a patients mother had a breast cancer gene. The Genos expression would be a yes or no value. But Genos may actually refer to the mothers entire gene bank, and perhaps her parents genes. A suitable browser would be able to explore these levels of information in ways that may not be predictable at the time the information was entered. A pre-Genos data base system that could work only with yes or no answers would still function normally. If a new understanding of genomic information emerged over time, Genos would be available for exploring it, and representing it with updated forms of display. Older, pre-Genos software may not be able to deal with the full richness of Genos expressiveness. In that case, these older programs would continue to use snapshot values from the M-Sheet. However, as new ways of looking at fractal growth patterns, gene expression and immune system interaction emerge, Genos would be able to adapt. Diachronic expressions are those describing things that change over time, a kind of time travel. In some cases, historical data may be discovered to have been incorrect. Other times, it may be that the nomenclatures and terminology change out of synchronization with fields that refer to them. Synchronic expressions are those taken at a snapshot in time. These two terms (diachronic and synchronic) come from linguistics, in which the diachronic study is the flow of a language over time (Olde English to current English, for example) and synchronic is the study of a language as it is used at an instant in time (such as would be used by the editor of a newspaper.) Many pre-Genos systems are not able to deal with diachronic information, or must do so only through elaborate application-level reporting schemes. Genos would be able to present a synchronic expression of a field for the M-Sheet, allowing pre-Genos systems to process the data, yet retain the diachronic information for examination by richer M-Sheet viewers. Many of the relationships expressed by Genos entail temporal relationships. Some can be expressed as a calendar-type date; others may be part of a cascade with separate timelines and identities to manage. Other characteristics that mark the concept of Genos include the following: Genos may allow inverted thinking. For example, it might express the perspective of a disease that is trying to survive within a population of humans hostile to its growth. The time and growth expressions inherent in Genos would use a complex array of timelines to express the development of the disease and its evolution. This information would appear on an individuals M-Sheet. A lab result may include some antibiotic resistance information at the M-Sheet level, compatible with systems as expressed today, but which could be further examined for population-based resistance and trends, as well as a predictive model to determine how the microbe might mutate to its greatest evolutionary advantage. Genos could also allow us to drill down on a particular microbe and find a broad array of information, such as its potential for tipping an epidemic, mutating to a resistant strain, or other behavior. This concept has some interesting implications for the role of identity and timelines in Genos. Is an entity defined by its timeline? What happens to the identity of an entity if it mutates over time? When does it assume a new identity? Genos would be able to express the synchronic value of the information, and still allow browsing for deeper information. Desired Traits of Genos We can show easily that Genos cannot have the simple precision and determinacy of Newtons clockwork universe model. Living, growing processes are not susceptible to the same precision as the inanimate world Newton described. For example, if we were to replace Newtons billiard balls with mice of the same mass, we could not expect the same simplistic behavior when we hit one with a cue stick. Genos will not be a simple as F=MA. It is instructive in attempting to conceptualize Genos to look at some aspects of Newtons approach versus that of Aristotle: Newton dealt directly with the void, or zero. Newton had to deal with the infinite, or infinity Newton had to jump out of Aristotles genus/species model, as well as the specific genera of the model. If we assume that current medical informatics models are indeed defined according to the Aristotelian model, then Genos might have the same hurdles to overcome that Newton did and perhaps even more of them. Dealing with the Void Since Aristotelian doctrine focused only on differentiating things that existed, we are unable to use it to think about things that do not exist, as Newton was able to do in his use of the concept of void. Decision makers who are presented with a prioritized list of problems that exist and have tangible costs are hard put to entertain discussion of approaches that do not address existing problems. This leads, for example, to shortsighted decisions regarding vaccination policy and other public health issues. Notions of trust, community, and self-efficacy that may have profound effects on the health care process are nothing to the disease categorization process. Starting with a genus of disease, we are blinded to relationships of health. In order to discuss this void in terms of Genos, we will introduce the notion of autocatalytic space. This is a void that grows as more material is added to it. An example of an autocatalytic space is the World Wide Web. It started out as simply a set of standards for HTML, HTTP, and URL. There was no content, just an empty space ready for accepting information. As more sites joined the web, the void grew larger, fueling itself. When Amazon.com appeared, it did not use up space on the web, but rather created a larger web with more users. Had it opened a bookstore in a shopping mall, however, it would have decreased the space available to for others to use. Autocatalytic space is not used up when stuff appears in it, but rather fuels greater acceptance of more. For example, someone elses immune system that successfully defeats an attack of tuberculosis improves others health. The healthier each person becomes, the healthier all can be. Health can thus be viewed as an autocatalytic space; it does not have to be viewed solely as a scarce commodity subject to depletion when it is consumed. Traditional economic models are based on the notion of scarcity, what may be called entropic space. In this model, patients consume health care and deplete a scarce resource. The health care system produces health care, and a complicated profit/insurance/risk/regulatory mechanism attaches a price to this exchange. Things costing nothing are not considered; they are cast into the void, when in reality, they may be the most valuable solutions of all. This market creates value based on disease and the consumption of healthcare. Metrics used to evaluate the value created by an investment are expressible only in terms of disease markers used in the nomenclature. Someones immune system fighting off tuberculosis is invisible to the metrics, and therefore, not valued. Autocatalytic values and cascades of effects are simply ignored by the present system. Autocatalytic spaces may also be applied to epidemiology, as is the case, for example, in the herd effect. The concept may be useful in addition for describing cascading activities in individual, evolutionary processes, as well as developmental biology and morphogenesis. The autocatalytic space model allows Genos expressions to value nothing. Dealing with the Infinite Things ubiquitous to a given genus/species/differentia often are inexpressible in breaking down that genus/species/differentia. For example, concepts such as an immune system or homeostasis affect practically all disease categories they are ubiquitous. But to express them in each specific disease categorization would be impossible. They would have to be put in a different axis or dimension, and each would end up with its own axis, creating an impossibly large number of dimensions. Such factors could be termed infinite relative to the specific deconstructions of disease. Similarly, the effects of genetic modifications on the evolution of the human species or of pathogens are not necessarily expressible within languages based on predictability and causality. Expressions in Genos may need to describe cascades that occur at several different levels simultaneously. For example, the overuse of antibiotics may cause a cascade of events that affects future diseases, health care financing, bioterrorism, and political decisions. Each of these may induce a cascade of further activity. These cascades, or avalanche effects, must be expressible in Genos. Jumping Out of the Model Newton could not have succeeded had he started by accepting Aristotles model. The genus of motion as circular and straight and the laws of straight motion could not have led him to develop F=MA. In the same way, Genos cannot start by assuming the current breakdown of disease, the snapshot perspective of categorizations, and the fixed nature of definitions as the starting point. It is possible that Genos may finally deconstruct back to the current model, but this is not a guaranteed result. For example, genomics may teach us that our diagnostic categories and procedures are completely at odds with the way the genes express themselves. Therefore, Genos must be agnostic with respect to any One Correct Way of organizing information, allowing expressions themselves to jump out as necessary. Some Properties of Genos From the above analogy, we can begin to describe some of the properties of this desired framework: Genos must have advanced security and privacy capabilities. Dealing with genomic information is far more challenging than with other types of clinical information. Patient identities can be discovered covertly. For example a blood sample sent with the expectation of being an anonymous test contains DNA information which could violate that anonymity. Genetic information which may have been given in an earlier era could yield confidential information as later technology progresses. Expressions of genomic information are much more sophisticated than simple field values contained in lab tests, for example. Genos would have to be able to successfully make refutable predictions. As Karl Popper points out, the vagueness of daily horoscopes makes them impossible to refute. Having an irrefutable theory is a measure of its weakness, not its strength. The criterion for scientific validity of the language would be its falsifiability, or refutability, or testability. This is a challenging problem, lest the system lapse into gibberish in the name of generality. The transformations Genos applies to its data to arrive at an M-Sheet expression for display must be visible to browsers, so that they can make their own judgments as to its validity. It is unlikely that Genos will be provably consistent, one of the stated goals of many medical informatics systems. Although clarity is valuable in itself, exactness or precision is not: there can be no point in trying to be more precise than our problem demands. Linguistic precision is a phantom. Genos must be able to deal with messy information that has meaning but does not fit Aristotelian deconstruction. Genos must be able to deal with the concept of belief systems. Peoples beliefs and expectations about themselves and the predictions of a given therapy have a significant effect on their health. Someone who truly believes that burying a potato at night will improve his or her health may indeed find that this is true. Regardless whether a term is statistically meaningful according to Gaussian distributions, it may have meaning in a specific context. Genos cannot ignore information merely because it is inconsistent. It must be able to build trust in its users, so that they are able to judge for themselves the source of information and the type of validity, if any, is claimed for expressions. The Genos framework deals with expressions (which may be evaluated to become traditional fields) and may therefore offer higher levels of security. Genos must be able to deal with various conundrums relating to diagnoses and discourse based on it. For example, Gergens progressive infirmity loop describes a situation in which a named illness begets specialists looking for it, which creates a diagnosis, creating a billing code, creating additional cases to be treated by specialists. Additionally, Genos must express media-driven medicine issues, and the role of the Internet and media in alleviating/creating illness and health. Publicity surrounding an illness can generate self-fulfilling predictions about the spread of the illness. The connectivity of the Internet and the publics increasing access to medical information (and medical superstition) online must be addressed as an intrinsic part of the Genos framework. Genos must be able to express the positive discourse concepts of health, rather than being locked in solely to the deficit discourse concepts of current nomenclatures. This is particularly critical with regard to our attempts to understand genomic information: It is possible that research could be directed solely towards the categorizations of disease we now recognize, resulting in reinforcement of current categorizations, rather than being open enough to detect fundamentally new relationships. This is not to deny the power of the disease-based model, but rather to ground it more firmly within a framework of health. Genos must be capable of self-reference. DNA is self-referential, in the sense that that it describes both the machine for interpreting DNA code and the code itself. As we progress in our understanding of genomics, we face many paradoxes of self-reference. Genos must deal with a variety of characteristic scales. We may view the domain of discourse to be a continuum of scale, from the smallest molecule or gene to cells, organs, bodies, families, community, nations, humanity, and the evolution of the species. Various sciences tend to focus on specific sections of this continuum, often narrowing their range of study in order to maintain the causality of the relations they study. Genos must be able to express the full continuum at multiple scales, which may have a cascade effect. Cascades may operate at multiple levels simultaneously. Current nomenclatures focus on a single characteristic scale diseases of a single person or body part in a clinical setting. Genos must be able to express the meanings of things that evolve over time. For example, issues of stem cell research deal with the evolution of the human species. This is a far more complex problem than that faced by current clinical information systems. Genos must be able to express evolution, human development, growth, and the change of meaning over time. Genos must be able to express self-organization. Things that organize themselves and create order may be most meaningfully understood by understanding their initial conditions, requirements for evolutionary growth, and constraints on their evolution, rather than a precise accounting at some snapshot in time. Genos must be able to describe fractal activities and patterns. This is a shift from what is implicit in traditional medical categorization, which assumes that more precise measurements will yield less error. For example, someone measuring a coffee cup with ever more precise rulers will arrive at decreasing errors of measurement, according to Gaussian distributions. But someone measuring a coastline with ever-smaller rulers will find that the length of the coastline becomes increasingly longer. The length of the coastline varies with the length of the ruler. Genos must be able to express both the precision of the Gaussian-described values, as well as the scale-dependent precision of fractal dimensional measurement. Fractal values are common in medical measurement, from the surface area of the brain to the rhythms of the heart. Genos must be able to express foamy spaces, rather than continuous spaces in fixed dimensions. For example, an information space may describe male and female as dichotomous properties of a person. However, a transgendered person may fit into both of these categories over time. This persons information space may be said to be foamy in that within one point, it contains all of the dimensions of either male or female. The pigeonhole paradigm assuming that information can be normalized into predefined rows and columns has served us well in some areas of computing, but is not necessarily expressive of underlying meaning. New approaches using eXtensible Markup Language, Semantic Webs, and Topic Maps, for example, open up new ways of expressing meaning and relationships beyond traditional relational database models. Genos must be able to express growth patterns and things that are grown rather than built. Full meaning of an organism is not understood by looking at a growth process from just a snapshot in time. Some things can only be expressed in terms of their growth: Wrinkled folds on the surface of the brain maximize surface area; the respiratory, lymphatic, and nervous systems all have a fractal treelike organization similar to the circulatory system; complicated fractal folding is found on the microscopic level of the kidneys and also on the microscopic level of mitochondria. Nature minimizes material requirements while maximizing functionality through the ubiquitous use of fractals. However, all of our examples have one very important aspect that, while obvious once seen, is easily overlooked. Fractals in biological systems must be grown. Why is this important? When something is grown, the instructions for how the growth is to progress must be contained somewhere. A framework such as Genos, if ever created, would probably as revolutionary to our understanding of health as Newtons concepts of the universe would have been to Aristotle. There is no guarantee that such a language could exist. The above wish list pulls in aspects of chaos theory, fractals, self-organization, information and negentropy, autocatalytic networks, global connectivity, and genomic information. These advances are just decades old, not long enough to be fully studied, possibly discredited, or become mature sciences. Where to go from here? It is simple to portray a hypothetical framework that solves all of our problems. In fact, it could be characterized as an irrefutable hypothesis. Creating that framework and validating it within a real-world environment is much more difficult, and would be subject to testability and refutation. Some future directions for exploring the viability of Genos may be: Consider the use of Genos as a framework for developing VHAs Health ePeople concepts. Survey existing genomic and bioinformatics research for emerging languages that may be used or adapted. Investigate the role of genomic information and the clinical information system, with an eye towards creating ways of incorporating genomic information in clinical information system in richer ways. Look at advances in genomic information as fundamental shifts in our understanding of clinical information, not necessarily something to be agglutinated to existing clinical knowledge bases. Investigate paths VHA can take today that will allow it to better adapt to genomic information in the future. Investigate ways in which complex health issues, such as bioterrorism, could be addressed with more sophisticated language. Conclusion As advances in medical science and genomics appear with ever increasing rapidity, it is essential that we not become Martians pulling tubes from televisions as we attempt to deal with and describe new systems, relationships and complexities. A framework such as Genos could help us embrace this exciting cascade of discovery with a focus on what is right about human health and at the same time could help us better express the relationship between its components of disease and the whole of the individual. A framework such as Genos provides a point of integration for future advances in health-oriented understanding genomics and proteomics and their clinical application. It also provides a foundation for a more advanced security system that is able to deal with the semantics of genomic information. Tuttle, Mark S., et al, The Semantic Web As Perfection Seeking: A View from Drug Terminology  HYPERLINK "http://www.semanticweb.org/SWWS/program/full/paper49.pdf" http://www.semanticweb.org/SWWS/program/full/paper49.pdf  Shortliffe, et al, Medical and Informatics, Computer Applications in Healthcare Biomedicine, Springer, 1999, p. 181  Campbell KE, Das AK, Musen MA. A Logical Foundation for Representation of Clinical Data. Journal of the American Medical Informatics Association 1994; 1(3):218-232.  Lipow SS, Campbell KE, Olson NE, et al. Formal Properties of the Methathesaurus: An Update. In: Gardner RM, ed. Proceedings of the Nineteenth Annual Symposium on Computer Applications in Medical Care. New Orleans: Hanley & Belfus, Inc., 1995:944.  Campbell, Keith, et al, Scalable Methodologies for Distributed Development of Logic-Based Convergent Medical Terminology  HYPERLINK "http://www.mayo.edu/imia-wg6/conf/doc/campbell.pdf" http://www.mayo.edu/imia-wg6/conf/doc/campbell.pdf  These comments are taken from a draft copy of a paper written for the National Initiative for Health Informatics (NIHI) Clinical Specificity/Content and Structure committee.  Russell, Bertrand, A History of Western Philosophy, Simon and Schuster, 1954, p. 73  Seife, Charles, Zero, The Biography of a Dangerous Idea, Penguin, 2000, pp 25, 46   HYPERLINK "http://www.rit.edu/~flwstv/aristotle1.html" http://www.rit.edu/~flwstv/aristotle1.html  Popper, Karl, Conjectures and Refutations, The Growth of Scientific Knowledge, Routledge and Kegan, plc, 1963,p. 8  Seife, Charles, Zero, The Biography of a Dangerous Idea, Penguin, 2000, p. 77  Madsen, I, and Burgaard, SA The Danish National Health Classification System,  HYPERLINK "http://www.nur.utexas.edu/0305/jkang/JKweb/Nurse/TERM.pdf" http://www.nur.utexas.edu/0305/jkang/JKweb/Nurse/TERM.pdf  Cimino, James, Desiderata for Controlled Medical Vocabularies in the Twenty-First Century  HYPERLINK "http://www.mayo.edu/imia-wg6/conf/doc/cimino.pdf" http://www.mayo.edu/imia-wg6/conf/doc/cimino.pdf  Jonas Salk asked himself what it would be like to be a polio virus, for example.  Popper, Karl, Conjectures and Refutations, The Growth of Scientific Knowledge, Routledge and Kegan, plc, 1963,p. 37  ibid, p. 28  This could be attributed to psychogenic causes, or simply the exercise of digging holes at night.  Gergen, Kenneth, Is Diagnosis a Disaster?  HYPERLINK "http://www.swarthmore.edu/SocSci/kgergen1/text5.html" http://www.swarthmore.edu/SocSci/kgergen1/text5.html  Liebovitch, Larry, Fractals and Chaos, Simplified for the Life Sciences, Oxford University Press, 1998 World Wide Web Consortium, XML activities:  HYPERLINK "http://www.w3c.org/XML" http://www.w3c.org/XML  World Wide Web Consortium, Semantic Web activities  HYPERLINK "http://www.w3c.org/2001/sw" http://www.w3c.org/2001/sw   HYPERLINK "http://www.topicmaps.net" http://www.topicmaps.net  Flake, Gary William, The Computational Beauty of Nature, Computer Explorations of Fractals, Chaos, Complex Systems, and Adaptation, Bradford Press, 1999, p. 77-78 PAGE  PAGE 1 & ' N O P b c d n o p {        : ; < = Y Z g\h4mHnHu2jh~Dh4>*B*UmHnHphuh4mHnHuh~Dh40JmHnHu$jh~Dh40JUmHnHu hf0Jjh0UjhfUh:|Lhf0JaJjhfCJU hfCJjhfCJU hf5CJ&hf hf5CJ$$,-A`  & d   {  9b9999999999979999999999979 !  ! $a$"e~Z [ u v w x y z { | }  𶭶t𶭶Z2jh~Dh4>*B*UmHnHphu jAh4UmHnHu2jh~Dh4>*B*UmHnHphuh~Dh40JmHnHuh4mHnHu$jh~Dh40JUmHnHuhQmHnHu jGh4UmHnHuh4mHnHujh4UmHnHu"         2 3 4 5 J K L f g h i j k l m n ¹««u¹«[«2jh~Dh4>*B*UmHnHphu j5h4UmHnHuh4mHnHu2jh~Dh4>*B*UmHnHphuh~Dh40JmHnHuh4mHnHu$jh~Dh40JUmHnHuhQmHnHujh4UmHnHu j;h4UmHnHu  l P 6 k_n'(*HIJ99|9999 $7$8$H$`a$ 7$8$H$` !  !  . / 0 J K L M N O P Q R n o p q ¹««u¹«[«2jh~Dh4>*B*UmHnHphu j)h4UmHnHuh4mHnHu2jh~Dh4>*B*UmHnHphuh~Dh40JmHnHuh4mHnHu$jh~Dh40JUmHnHuhQmHnHujh4UmHnHu j/h4UmHnHu    0 1 2 3 4 5 6 7 8 T U V W z { | ¹««u¹«[«2j h~Dh4>*B*UmHnHphu j h4UmHnHuh4mHnHu2jh~Dh4>*B*UmHnHphuh~Dh40JmHnHuh4mHnHu$jh~Dh40JUmHnHuhQmHnHujh4UmHnHu j#h4UmHnHu IJKe¹««u¹«[«2j h~Dh4>*B*UmHnHphu j h4UmHnHuh4mHnHu2j h~Dh4>*B*UmHnHphuh~Dh40JmHnHuh4mHnHu$jh~Dh40JUmHnHuhQmHnHujh4UmHnHu j h4UmHnHu efghijklm=>?Y¹««u¹«[«2j h~Dh4>*B*UmHnHphu j h4UmHnHuh4mHnHu2j h~Dh4>*B*UmHnHphuh~Dh40JmHnHuh4mHnHu$jh~Dh40JUmHnHuhQmHnHujh4UmHnHu j h4UmHnHu YZ[\]^_`a}~¹««u¹«[«2jvh~Dh4>*B*UmHnHphu jh4UmHnHuh4mHnHu2j|h~Dh4>*B*UmHnHphuh~Dh40JmHnHuh4mHnHu$jh~Dh40JUmHnHuhQmHnHujh4UmHnHu j h4UmHnHu 4567LMNhijklmnop¹««u¹«[«2jjh~Dh4>*B*UmHnHphu jh4UmHnHuh4mHnHu2jph~Dh4>*B*UmHnHphuh~Dh40JmHnHuh4mHnHu$jh~Dh40JUmHnHuhQmHnHujh4UmHnHu jh4UmHnHu !"#$%&'()EFGH`ab|¹««u¹«[«2j^h~Dh4>*B*UmHnHphu jh4UmHnHuh4mHnHu2jdh~Dh4>*B*UmHnHphuh~Dh40JmHnHuh4mHnHu$jh~Dh40JUmHnHuhQmHnHujh4UmHnHu jh4UmHnHu |}~"¹««u¹«[«2jRh~Dh4>*B*UmHnHphu jh4UmHnHuh4mHnHu2jXh~Dh4>*B*UmHnHphuh~Dh40JmHnHuh4mHnHu$jh~Dh40JUmHnHuhQmHnHujh4UmHnHu jh4UmHnHu "#$%&'()c h !!# #,#0#h#p###r$s$((((D-I-k-p-H;I;<<X<Y<<<<<===?>?@@tAAA¹~hfB*ph jhf0JB*CJUphjhf0JU hf6 hf6]hfjhfUh4mHnHu$jh~Dh40JUmHnHuhQmHnHujh4UmHnHu jh4UmHnHu0uTZ!"##$$%%.%>%e%%%9999999999999999999$9$9$9$9$99$ & F 7$8$H$ & F 7$8$H$ 7$8$H$`gd:|L%% &|&&&3(M*,- /012V46-9;I;J;[<==??K@A9$9$9999999999|999999999x9x999 7$8$H$`gd` & F 7$8$H$AAAOEPEKKtLuLMM W WWW^^^^ooqqttv vfvpvuvvzz,1Tg @Iz{œΜĝŝ{|_pqƳdz?F hfCJh7ho<h hf6] hf6jhf0JUhfjhf0JB*Uphhf0JB*phIABBRESE FFFF G.HIJKM,NNNOQRTTWYY999x9x999x9x99t;99t;9999999999999`h`h & F 7$8$H$`YZ ]^accdfh6klmoqstxyz{|'݀\999999999999|99999999n9n90 999 & F  & F gd``gd\]'(g5Ѝ.Hޓ'@999999999999999999999990 99 gd & F` & F ^9:!"()ef>?FG99n99϶99 990 9999 99n9999϶99϶999x9x9x9x9x^ & F & F4KLx}Dq!"~%9x999|9999999999999997$8$H$gd7 & F `h^h^V[!"# e~%&'*Snopqپٺٰٰjh4U h40Jjh4Ujh4U h45jh40JUh4jLh4CJUaJjh4CJUaJh4CJaJjh40JCJUaJhE hf6h7hf5nMsB }'ydenop{|}~9&`#$'MN]567pqst@ABC }~$%'(<p12Zjh4Ujh4Ujdh4Ujh4U h45jh40JUh4 h40Jjh4UBZ[\vwyz{|Cdeflmnpqwxyz{}ĺ⺴hQ0JmHnHu h40Jjh40JUhf h45jh4Ujh40JUh4 h40Jjh4Ujh4U 1h/ =!"#$%DyK munnecket@saic.comyK 4mailto:munnecket@saic.comDyK FlC:\Documents and Settings\Administrator\Local Settings\Local Settings\Temporary Internet Files\OLK7\D21.doc}DyK _Toc532805134}DyK _Toc532805134}DyK _Toc532805135}DyK _Toc532805135}DyK _Toc532805136}DyK _Toc532805136}DyK _Toc532805137}DyK _Toc532805137}DyK _Toc532805138}DyK _Toc532805138}DyK _Toc532805139}DyK _Toc532805139}DyK _Toc532805140}DyK _Toc532805140}DyK _Toc532805141}DyK _Toc532805141}DyK _Toc532805142}DyK _Toc532805142}DyK _Toc532805143}DyK _Toc532805143}DyK _Toc532805144}DyK _Toc532805144}DyK _Toc532805145}DyK _Toc532805145}DyK _Toc532805146}DyK _Toc532805146}DyK _Toc532805147}DyK _Toc532805147}DyK _Toc532805148}DyK _Toc532805148}DyK _Toc532805149}DyK _Toc532805149}DyK _Toc532805150}DyK _Toc532805150}DyK _Toc532805151}DyK _Toc532805151}DyK _Toc532805152}DyK _Toc532805152}DyK _Toc532805153}DyK _Toc532805153}DyK _Toc532805154}DyK _Toc532805154YDyK 9http://www.semanticweb.org/SWWS/program/full/paper49.pdfyK rhttp://www.semanticweb.org/SWWS/program/full/paper49.pdfADyK 3http://www.mayo.edu/imia-wg6/conf/doc/campbell.pdfyK fhttp://www.mayo.edu/imia-wg6/conf/doc/campbell.pdf!DyK +http://www.rit.edu/~flwstv/aristotle1.htmlyK Vhttp://www.rit.edu/~flwstv/aristotle1.html]DyK :http://www.nur.utexas.edu/0305/jkang/JKweb/Nurse/TERM.pdfyK thttp://www.nur.utexas.edu/0305/jkang/JKweb/Nurse/TERM.pdf9DyK 1http://www.mayo.edu/imia-wg6/conf/doc/cimino.pdfyK bhttp://www.mayo.edu/imia-wg6/conf/doc/cimino.pdfIDyK 5http://www.swarthmore.edu/SocSci/kgergen1/text5.htmlyK jhttp://www.swarthmore.edu/SocSci/kgergen1/text5.htmlDyK http://www.w3c.org/XMLyK .http://www.w3c.org/XMLDyK http://www.w3c.org/2001/swyK 6http://www.w3c.org/2001/swDyK http://www.topicmaps.netyK 4http://www.topicmaps.net/@@@ NormalCJ_HaJmH sH tH Z@Z Heading 1$<@&5CJ KH OJQJ\^JaJ \@\ Heading 2$<@& 56CJOJQJ\]^JaJV@V Heading 3$<@&5CJOJQJ\^JaJX@X Heading 4$7$8$@&H$`5CJ OJQJ]N@N Q Heading 5 <@&56CJ\]aJH@H Q Heading 6 <@&5CJ\aJ:@: Q Heading 7 <@&@@@ Q Heading 8 <@&6]N @N Q Heading 9 <@&CJOJQJ^JaJDA@D Default Paragraph FontVi@V  Table Normal :V 44 la (k@(No List >@>  Footnote TextCJaJ@&@@ Footnote ReferenceH*FOF Quotexx]^m$CJ4U@!4  Hyperlink >*phFV@1F FollowedHyperlink >*B* phDC@BD Body Text Indent ^.@. TOC 2 ^.@. TOC 3 ^H@rH  Balloon TextCJOJQJ^JaJFOF quote]^CJaJL^@L Normal (Web)dd[$\$ B*ph33fbOb Heading 1 Char.5CJ KH OJQJ\^J_HaJ mH sH tH RY@R  Document Map-D M OJQJ^J&@& TOC 14 @4 Footer  !.)@. Page NumberXR@X Body Text Indent 27$8$H$`6]DT@D Q Block Text x]^2B@2 Q Body Text!x<P@"< Q Body Text 2 "dx>Q@2> Q Body Text 3#xCJaJPM@BP QBody Text First Indent $``N@AR` QBody Text First Indent 2%hx^h`TS@bT QBody Text Indent 3&hx^hCJaJ@"@@ QCaption 'xx5CJ\aJ2?@2 QClosing (^<@< Q Comment Text)CJaJ@j@@ QComment Subject*5\$L@$ QDate+<[@< QE-mail Signature,<+@< Q Endnote Text-CJaJh$@h QEnvelope Address!.@ &+D/^@ OJQJ^JN%@N QEnvelope Return/CJOJQJ^JaJ4@4 QHeader 0 !:`@: Q HTML Address16]Re@"R QHTML Preformatted2CJOJQJ^JaJ: @: QIndex 13^`: @: QIndex 24^`: @: QIndex 35^`: @: QIndex 46^`:@: QIndex 57^`:@: QIndex 68^`:@: QIndex 79^`:@: QIndex 8:^`:@: QIndex 9;p^p`H!@2H Q Index Heading<5OJQJ\^J4/@4 QList=h^h`82@8 QList 2>^`83@8 QList 3?8^8`84@8 QList 4@^`85@8 QList 5A^`:0@": Q List Bullet B & F>6@2> Q List Bullet 2 C & F>7@B> Q List Bullet 3 D & F>8@R> Q List Bullet 4 E & F>9@b> Q List Bullet 5 F & FBD@rB Q List ContinueGhx^hFE@F QList Continue 2Hx^FF@F QList Continue 3I8x^8FG@F QList Continue 4Jx^FH@F QList Continue 5Kx^:1@: Q List Number L & F>:@> Q List Number 2 M & F>;@> Q List Number 3 N & F><@> Q List Number 4 O & F>=@> Q List Number 5 P & Fl-@l Q Macro Text"Q  ` @ OJQJ^J_HmH sH tH I@" QMessage HeadergR8$d%d&d'd-DM NOPQ^8` OJQJ^J>@2> Q Normal Indent S^4O@4 Q Note HeadingTDZ@RD Q Plain TextUCJOJQJ^JaJ0K@0 Q SalutationV6@@r6 Q Signature W^FJ@F QSubtitleX$<@&a$ OJQJ^J@ QTable 3D effects 1m:VYj#j#j#j#j.j.j.j. Y55\5B* \`J phB* `J ph6@6 QTable 3D effects 2:VZj.@j#j9jj 4Z5\5\t@t QTable 3D effects 3:V[j.@j j j#j9jj44[:B*`Jph B*`Jph5\5\br@b QTable Classic 1:V\0  j#j#j#jj \9B*`Jph6]5\56\]s@ QTable Classic 2#:V]0  j% j#j0 jjj%  ]:5\B*`JphB* `J ph5\Tt@T QTable Classic 3:V^0    jj0  j0  ^QB* ph5B*\`JphB* `J ph56B*\]`Jphu@ QTable Classic 4:V_0  jj0 j0 jj _X5\B* `J ph56B*\]`JphB* `J ph5\Rv@R QTable Colorful 1:V`0    j% j% jj%  `<B*ph56\]56\]56\]Tw@T QTable Colorful 2:Va0 j% jj0  j a@56\]56B*\]`Jph56\]4x@#4 QTable Colorful 3:Vb0j;$ j0 j%  b5B*\`Jphy@3 QTable Columns 1 :Vc0    j j jjjj#jj4cl5\B*`Jph B*`Jph5\5\5\5\5\5\z@C QTable Columns 2:Vdj j jjjj% jj4d5\B*`Jph B*`Jph5\5B*\`Jph5\B*`Jph5\5\{@S QTable Columns 3:Ve0j j jjj#j% j4eh5\B*`Jph B*`Jph5\5\5\B*`Jph5\|@c QTable Columns 4:Vfj j jjj% 4fLB*`Jph B*`Jph5\5\B*`Jph`}@s` QTable Columns 5:Vg0    j jjj#j#4gVB*`Jph B*`Jph5\5\5\56\]0@0 QTable Contemporary:Vh0j%@ j% j% 4h<@B*`JphB*`Jph5B*\`Jph@ Q Table Elegant_:Vi0j i;B*`Jphj@j Q Table Grid7:Vj0j~@ Q Table Grid 1z:Vk0jj k6]6]@ Q Table Grid 2:Vl0jjj#j l,5\5\5\5\@ Q Table Grid 3:Vm0  jjj0  m5\5\>@> Q Table Grid 4:Vn0  jj0 j0  nB5B*\`Jph5B*\`JphB*`Jph@ Q Table Grid 5:Vo0    jjj# j o5\5\$@$ Q Table Grid 6:Vp0    jj#j#j p(5\B*`Jph5\J@J Q Table Grid 7:Vq0    jjj#j# j q25\5\5\5\5\ @#  Q Table Grid 8:Vr0jjj%  rH5B*\`Jph5B*\`Jph5B*\`Jphv@3v Q Table List 1:Vs0  j%@ jj#j0 j4sM@B*`JphB*`Jph56B* \]`J ph5\p@Cp Q Table List 2:Vt0 j%@ jj#j0 j4tG@B*`JphB*`Jph5B*\`Jph5\@S Q Table List 3:Vu0  j# j# j u05B* \`J ph6B* ]`J ph@c Q Table List 4w:Vv0    j0   v5B*\`Jph@s Q Table List 5:Vw0jj#  w5\5\@ Q Table List 6:Vx0j%@ j# j#  4x5\5\@ Q Table List 7:Vy0  j%@ j% jjj# j0  4y>@B*`Jph5\5\5\5\@ Q Table List 8:Vz0j%@ j% jjj#j0 4zD@B*`Jph5\5\5\56\]T,@T QTable of Authorities{^`L#@L QTable of Figures| ^` @ QTable Professionall:V}0j%  }5B*\`Jpho@ QTable Simple 1:V~0  j#j# ~rp@r QTable Simple 2:Vj#j# j#j# j#j#O5\5\5B*\`Jph5\5\5\q@ QTable Simple 3l:V0    j%  5B*\`Jphp@p QTable Subtle 1 :Vj0@ j# j# j0  j. jj45\5\p@#p QTable Subtle 2:V0j0  j0  j# j# jj5\5\l@3l Q Table Theme7:V0@C Q Table Web 1h:V03j B*`Jph@S Q Table Web 2h:V03j B*`Jph@c Q Table Web 3h:V03j B*`JphV>@rV QTitle$<@&a$5CJ KHOJQJ\^JaJ H.@H Q TOA Headingx5OJQJ\^J.@. QTOC 4 ^.@. QTOC 5 ^.@. QTOC 6 ^.@. QTOC 7 ^.@. QTOC 8 ^.@. QTOC 9 ^r 4X45=79O=CtDE OOrzĕ{ƫ \L+{Q s[  n W B E  E   | ՟ b . _ uF I۞ Gm o K8Qp ,-A`&d{lP6k_n' ( * H I J uTZ.>e |3 M"$% '()*V,.-13I3J3[455?7K89::R=S= >>>> ?.@ABCE,FFFGIJLLOQQR UVY[[\^`6cdegiklpqrstw'wx\y]yyy'z(zz{}~g5Ѕ.Hދ'@9:!"()ef>?FG4KLx}Dq!"~%߼ľnؿMsB }'ydenop{|}0000000000000000000000000p000000000000000000p00000* 0* 0* 0* (0* 0 0 (0* 0 0 (0* 00000000(0* 00 0 0 0 0 00 0 0 00(0* 0|0|0|0|0|0|0|00(0(0(0(0((0(03030303030303(0(09090909090909 09 09 09 09 09090909(0(0F0F0F0F(0(0L0L(0(0Q0Q0Q0Q0Q(0(0[0[0[0[0[00d0d0d0d0d0d0d 0d 0d 0d(0d0w0w0w 0w 0w 0w0w0w(0d0z0z0z0z0z0z0z(0d0.0.(0d0ދ(0d0'0' 0'0' 0'0' 0'0' 0'0' 0'0' 0'0' 0'0' 0'0' 0'0' 0 '0' 0 '0' 0 '0' 0 '0'0'0'0'0'0'0'0'004040404 04 04 04 04 04 040<(0<00000000000000000000000000000000000 Z  eY|"AZhklnopqrstuvy %AY\imwxz{|}~~j&Ob<Zvxy{4Kgijl/KMNPp1346V{Jfhik>Z\]_6Mikln " $ % ' G a }   # % & ( XX X%X%X%X%X%X%X%X%X%X%X%X%X%X%X%X%X%X%X%X%X%̕ !!bzNN   9 T Y E XXXXXXXXX8@0(  B S  ? _Toc532805134 _Toc532805135 _Toc532805136 _Toc532805137 _Toc532805138 _Toc532805139 _Toc532805140 _Toc532805141 _Toc532805142 _Toc532805143 _Toc532805144 _Toc531683301 _Toc532805145 _Toc532805146 _Toc532805147 _Toc532805148 _Toc532805149 _Toc532805150 _Toc532805151 _Toc532805152 _Toc532805153 _Toc532805154 _Toc530553778+ |(39FLQQ[dwz.ދ'4h G )H3:FLQ[[d&w{G?J!Hws ,uxsğ ysF zs}{sO |s}sd}~ss<_sps/s s sl s s st s]wslsd8ss,>s srsdrs<qs s, s@ses| |sv sE s slssss< sĂvs f s ssD^s/es5s<5sԞ s sqslqs,qsqs_s|_sDos=s$ls4qsesxslF s s<7hsl s sD sts s~s< sV ss\w--p kFkFFF}H}HTT[[\\__zwzwww1x1x0y0y]y]yyyyyzz{{>>DrryE      !"#$%&'()*+,-./012345678:9;=<>?@ABDCEFG77   qFqFFFHHTT[[\\__wwxx7x7x6y6ycycyyyyyzz{{͎͎CFFxM  !"#$%&'()*+,-./012345678:9;=<>?@ABDCEFG =!9*urn:schemas-microsoft-com:office:smarttags PlaceType="8*urn:schemas-microsoft-com:office:smarttags PlaceName:2E*urn:schemas-microsoft-com:office:smarttagsStreetB;*urn:schemas-microsoft-com:office:smarttagscountry-region9+B*urn:schemas-microsoft-com:office:smarttagsState>H*urn:schemas-microsoft-com:office:smarttags PersonName;3F*urn:schemas-microsoft-com:office:smarttagsaddress8=*urn:schemas-microsoft-com:office:smarttagsdate8FG*urn:schemas-microsoft-com:office:smarttagsCity9GH*urn:schemas-microsoft-com:office:smarttagsplace 122001DayMonthYearGF32GF+32GFGG!"GFGFGF32GFGFGFGFGFGFGFGFGFGFGFGFGF32GFGFGF+GF+GFGFG"!HG.4J4W4.P]Ml % a  & !"~ľnnpֿؿؿqssB }}%''dempz}"ezappjzappjzappjzappjzappjzappjzappj Tom Munneckezappjzappj|䆕P}*tBO~pN)"MF. Ex\DfCbL&(Bj w*8Cdg&x;0gj_p)x;0gf)`EmHnN{Kx;0gSVN8I;2T )TVP0a[{ax;0gG;9ox;0g^`.^`.88^8`.^`. ^`OJQJo( ^`OJQJo( 88^8`OJQJo( ^`OJQJo(hh^h`. hh^h`OJQJo(h ^`OJQJo(h pp^p`OJQJo(oh @ @ ^@ `OJQJo(h ^`OJQJo(h ^`OJQJo(oh ^`OJQJo(h ^`OJQJo(h PP^P`OJQJo(oh   ^ `OJQJo(h ^`OJQJo(h pp^p`OJQJo(oh @ @ ^@ `OJQJo(h ^`OJQJo(h ^`OJQJo(oh ^`OJQJo(h ^`OJQJo(h PP^P`OJQJo(oh   ^ `OJQJo(^`CJOJQJo(^`o(.pp^p`CJOJQJo(@ @ ^@ `CJOJQJo(^`CJOJQJo(^`CJOJQJo(^`CJOJQJo(^`CJOJQJo(PP^P`CJOJQJo(^`CJOJQJo(^`o(.pp^p`CJOJQJo(@ @ ^@ `CJOJQJo(^`CJOJQJo(^`CJOJQJo(^`CJOJQJo(^`CJOJQJo(PP^P`CJOJQJo(h^`OJQJo(hHhpp^p`OJQJ^Jo(hHoh@ @ ^@ `OJQJo(hHh^`OJQJo(hHh^`OJQJ^Jo(hHoh^`OJQJo(hHh^`OJQJo(hHhPP^P`OJQJ^Jo(hHoh  ^ `OJQJo(hHh^`OJQJo(hHh^`OJQJ^Jo(hHohpp^p`OJQJo(hHh@ @ ^@ `OJQJo(hHh^`OJQJ^Jo(hHoh^`OJQJo(hHh^`OJQJo(hHh^`OJQJ^Jo(hHohPP^P`OJQJo(hH^`CJOJQJo(^`o(.pp^p`CJOJQJo(@ @ ^@ `CJOJQJo(^`CJOJQJo(^`CJOJQJo(^`CJOJQJo(^`CJOJQJo(PP^P`CJOJQJo(^`o(. ^`hH. pLp^p`LhH. @ @ ^@ `hH. ^`hH. L^`LhH. ^`hH. ^`hH. PLP^P`LhH.hh^h`.h^`OJQJo(hHhpp^p`OJQJ^Jo(hHoh@ @ ^@ `OJQJo(hHh^`OJQJo(hHh^`OJQJ^Jo(hHoh^`OJQJo(hHh^`OJQJo(hHhPP^P`OJQJ^Jo(hHoh  ^ `OJQJo(hHh^`OJQJo(hHhpp^p`OJQJ^Jo(hHoh@ @ ^@ `OJQJo(hHh^`OJQJo(hHh^`OJQJ^Jo(hHoh^`OJQJo(hHh^`OJQJo(hHhPP^P`OJQJ^Jo(hHoh  ^ `OJQJo(hH^`CJOJQJo(^`o(.pp^p`CJOJQJo(@ @ ^@ `CJOJQJo(^`CJOJQJo(^`CJOJQJo(^`CJOJQJo(^`CJOJQJo(PP^P`CJOJQJo(^`CJOJQJo(^`o(.pp^p`CJOJQJo(@ @ ^@ `CJOJQJo(^`CJOJQJo(^`CJOJQJo(^`CJOJQJo(^`CJOJQJo(PP^P`CJOJQJo(;2TN{KmHG;9o{aCdg&j_p)0aSVNwj )TVf)`E~}|                                                                 o<:|LA`f7E40Q) (E`"ľnMsB mmmm@TJ@Unknowngz Times New RomanTimes New Roman5Symbol3& z Arial5& z!Tahoma?5 z Courier New;Wingdings"qhc\fc\fc\f[ _Q[ _Q%4dùù 3QL?:|LTowards a Language of Health Tom Munnecke Tom Munneckeh                  Oh+'0 0< X d p |Towards a Language of Healthowa Tom Munneckeguaom om  Normal.dote Tom Munneckegua3m Microsoft Word 10.0@Ik@ d@d@$ye [՜.+,D՜.+,H hp|  iQ_ùA Towards a Language of Health TitleDH  #; _PID_HLINKS_AdHocReviewCycleID_EmailSubject _AuthorEmail_AuthorEmailDisplayName_PreviousAdHocReviewCycleIDA 6_Toc5328051546z_Toc5328051536t_Toc5328051526n_Toc5328051516h_Toc5328051506b_Toc5328051496\_Toc5328051486V_Toc5328051476P_Toc5328051466J_Toc5328051456D_Toc5328051446>_Toc53280514368_Toc53280514262_Toc5328051416,_Toc5328051406&_Toc5328051396 _Toc5328051386_Toc5328051376_Toc5328051366_Toc5328051356_Toc532805134Y=lC:\Documents and Settings\Administrator\Local Settings\Local Settings\Temporary Internet Files\OLK7\D21.docv@mailto:munnecket@saic.comDhttp://www.topicmaps.net/bmhttp://www.w3c.org/2001/sw}#http://www.w3c.org/XMLM5http://www.swarthmore.edu/SocSci/kgergen1/text5.html  1http://www.mayo.edu/imia-wg6/conf/doc/cimino.pdf :http://www.nur.utexas.edu/0305/jkang/JKweb/Nurse/TERM.pdf q+http://www.rit.edu/~flwstv/aristotle1.htmlo{3http://www.mayo.edu/imia-wg6/conf/doc/campbell.pdf9http://www.semanticweb.org/SWWS/program/full/paper49.pdfqk Here it isHtom@munnecke.comiew Tom Munneckecom g  !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~Root Entry F&.eData s 1TablerWordDocument"SummaryInformation(DocumentSummaryInformation8CompObjj  FMicrosoft Word Document MSWordDocWord.Document.89q