[1]                           home                           [1]


Wednesday, December 07, 2005


 The BCNGroup Beadgames Index




Emergency Medical Ontology Project,

To be Proposed[PC1] 


Dr Jayanth G Paraki and Dr Paul S Prueitt

Draft: open to reading by third parties

Open to inclusion of third parties as co-sponsors


Draft: December 7, 2005




While existing representations of the biomedical domain may be sufficient for information retrieval[PC2]  purposes, the organization of knowledge in these representations is generally not suitable for reasoning by computers. Reasoning by computers requires the principled, consistent organization usually provided by ontologies.


An ontology can be viewed as a declarative model of a domain[PC3]  that defines and represents the concepts existing in that domain, their attributes, and the relationships between them. Knowledge engineers typically represent a constructed ontology as a “knowledge base”, though this phase has other meaning; such as the “knowledge base” within a corporation.  The constructed ontology becomes available to computer applications that need to use and/or share the knowledge of a domain, either with other computer programs or with humans. Within health informatics, ontology is a formal description of a health-related domain.


Health Informatics is a growing body of knowledge shared freely within third world countries.  Sharing is primarily within and between medical institutions, whose identification is part of the Emergancy Medical Ontology Project (EMOP) scope.  An ontology mediated information interface is proposed between an emerging third world Health Informatics (HI) knowledge base and medical science, as practiced in the US, India and Europe. 

Current scenario

“In recent years the development of ontologies—explicit formal specifications of the terms in the domain and relations among them (Gruber 1993)—has been moving from the realm of Artificial-Intelligence laboratories to the desktops[PC4]  of domain experts. Ontologies have become common on the World-Wide Web. Many disciplines now develop standardized ontologies that domain experts can use to share and annotate information in their fields. Medicine, for example, has produced large, standardized, structured vocabularies such as snomed (Price and Spackman 2000) and the semantic network of the Unified Medical Language System (Humphreys and Lindberg 1993). Broad general-purpose ontologies[PC5]  are emerging as well. For example, the United Nations Development Program and Dun & Bradstreet combined their efforts[PC6]  to develop the UNSPSC ontology which provides terminology[PC7]  for products and services”

Current work on medical ontologies

·        GALEN and the "Galen-Core" high-level ontology for medicine.


·        The ONIONS methodology - designed to build the ON9 medical ontology.


·        MedO - a bio-medical ontology developed at the Institute of Formal Ontology and Medical Information Systems, Germany.


·        TAMBIS (Transparent Access to Multiple Bioinformatics Information Sources) which uses an ontology of bioinformatics tasks and molecular biology to form a common user interface over multiple bioinformatics information resources.


·        The ontology for the HL7 Reference Information Model (RIM)

·        The Foundational Model of Anatomy - a domain ontology that represents a coherent body of explicit declarative knowledge about human anatomy.

What is needed?


What is urgently needed is a single robust ontology coupled to a decision support system[PC8] , which is capable of accepting real-time data, performing the necessary calculations, and providing a rapid output so that medical specialists can make the necessary decisions.

Do we have such a system[PC9] ?

Research on ontology is becoming increasingly widespread in the computer science community[PC10] . While this term has been rather confined to the philosophical sphere in the past, it is now gaining a specific role in areas such as Artificial Intelligence, Computational Linguistics, and Databases. Its importance has been recognized in fields as diverse as knowledge engineering, knowledge representation, qualitative modeling, language engineering, database design[PC11] , information integration, object-oriented analysis, information retrieval and extraction, knowledge management and organization, agent-based systems design. Current applications areas are disparate, including enterprise integration, natural language translation, medicine[PC12] , mechanical engineering, electronic commerce, geographic information systems, legal information systems, and biological information systems. Various workshops addressing the engineering aspects of ontology have been held in the recent years.


However, ontology by 'its very nature' ought to be a unifying discipline[PC13] . So far, the ontology discipline has not found common grounds. 


Insights in this field have potential impact on the whole area of information systems (taking this term in its broadest sense), as testified by the interest recently shown by international standards organizations[1]. In order to provide a solid general foundation for this work, it is therefore important to focus on the common scientific principles and open problems arising from current tools, methodologies, and applications of ontology.


There has been no modern in-depth analysis into medical ontologies despite the explosive developments in recent times.  The Emergancy Medical Ontology Project (EMOP) scope includes the development of foundational ontological principles as exposed by primary research into metabolic processes, cellular generation, immune system functioning and cognitive functioning.  These principles will be focused on near term development of stable Health Informatics ontology.  Foundational principles such as response degeneracy (as discussed by G. Edelman), many to many structure – function mappings in metabolic activity chains, results from Qualitative Structure Activity Relationship (QSAR) analysis, double articulation in emergent expression (such as in linguistics and in immunological response), and related principles.  The data construction we will use will be standardized on a generalization of the RDF triple, form a < subject verb predicate > triple to a more general “syntagmatic” < a, r, b > used extensively in graph theory.  The data encoding will standardize the hash table construction, as seen in the Berkeley hash table management system. 


Read write into and from RDF, Topic Maps, and OWL constructions will be mandated. 


Interoperability, plug and play, cooperative work, unambiguous communication, knowledge sharing, web environments[2]: all of these are examples both empowered and widely tackled by the current technology evolutions[PC14] .  However, we are focused in creating stable, easy to use conversion tools; while also extending the philosophi8cal understanding of the unique elements involved when one moves form purely engineering domains (such as software development), to the life sciences. 


The unambiguous communication of complex and detailed medical concepts[PC15]  is a problem still unresolved at the formal level. This situation has a significant impact on a very practical level with regards to medical information systems. The pitfalls of managing ambiguous[PC16]  communication within hospital information systems, for example, can have costly consequences[PC17] . Thus, the use of appropriate, up-to-date ontologies make for more effective data and knowledge sharing within Medicine, thus benefiting patients while maximising resources. These ontology constructions must be exportable into data structure that medical personal and the common man can understand. 




[1] ISO Technical Plan for Developing Nations

[2] World Community Grid

 [PC1]Prof Paul to add /subtract content appropriately to the document to make it complete and relevant to the tasks at hand

 [PC2]I am a Member of the Information Retrieval Group, Cochrane Collaboration, Cardiff, UK

 [PC3]Medical Virology and Immunology is the domain that addresses all issues related to the flu pandemic, HIV/AIDS, etc

 [PC4]How many doctors in USA have desktops in their offices?

 [PC5]Better to develop a holistic ontology

 [PC6]Grid Computing is the answer to contain spread of the flu virus

 [PC7]Terminology facilitates learning while Ontology facilitates pro-active decision making and results in a specific course of action

 [PC8]I like to work in a DSS Lab environment

 [PC9]Deming speaks of a system having 4 parts

 [PC10]Is there a SIG devoted to Medical Informatics?

 [PC11]Is there access to databases of flu epidemics?

 [PC12]Consider developing killer applications in Emergency Medicine

 [PC13]What constitutes unity? teamwork? harmony? peace? happiness?

 [PC14]Computers in Biology and Medicine, 2005

 [PC15]Let’s take into account only those concepts that impact on patient survival. We will call it the Survive to Live concept or Survive to Live Ontology. It can then reside on as many desktops as possible ….or laptops…

 [PC16]occurs when information of poor value is used in decision making and enforcing judgments

 [PC17]The costliest consequence occurs when a life threatening disease is diagnosed and yet the patient is left to die for want of timely treatment