home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
nlmpubs.nlm.nih.gov
/
2014.05.nlmpubs.nlm.nih.gov.tar
/
nlmpubs.nlm.nih.gov
/
umls
/
ism.txt
< prev
next >
Wrap
Text File
|
1994-04-25
|
14KB
|
465 lines
5. UMLS Information Sources Map
5.1. Introduction
The Information Sources Map (ISM) is a component of the Unified
Medical Language System (UMLS) project at the National Library of
Medicine (NLM). It is a knowledge source which is intended to
draw upon the capabilities of the the UMLS Metathesaurus and UMLS
Semantic Network, with two chief aims:
1) To describe electronically available information sources in
machine-readable form in such a way as to allow a user of
appropriate computer software to determine which sources are
appropriate to a particular query. We refer to this as the
descriptive component of the ISM. The prototype ISM file
supplied on this disk contains data on the major publicly
available databases of the NLM, as well as databases and
expert systems programs from outside of the NLM dealing in
human genetics, medical diagnosis, and environmental health
and toxicology.
2) To provide requisite information to allow automatic
connection to the information source, and the completion of
a successful information retrieval session. We refer to
this as the procedural component of the ISM. Procedural
tools for the ISM are under active development, employing
recently developed techniques and standards for network
information retrieval; it is hoped that some of this
software will appear on a future UMLS CD-ROM.
5.2. Format of ISM data
The data exchange format for the Information Sources Map is an
ASCII representation of a set of relational tables; it is
expected that the information in these tables will be transformed
into other representations appropriate for use by specific
experimenters and developers. A single table equates to a DOS
text file in the MSDOS environment.
The ISM tables are:
ISMR Main Information Sources Map records
ISDEFR Field definitions for main records
ISVALR Valid values for ISM subfields which use a
controlled vocabulary
5.3. Descriptions of ISM Tables
5.3.1. ISMR (Main Information Sources Map Records)
Each field (row) in this table defines a particular item of
information concerning an information source, and is organized
April 22, 1993
into subfields as follows:
Subfield 1: Unique identifier of the ISM record.
Subfield 2: Field identifier from the set defined in ISDEFR.
Subfields 3-n: Subfields following the format defined in ISDEFR,
with a vertical bar separator (|, ASCII 174)
between subfields.
A logical record in ISMR is defined as all table fields (rows)
which share a common unique identifier in subfield 1. An ISM
main record requires at least the presence of an entity name
(subfield 2: NA_) combined with a provider (subfield 2: PO_).
Although there should generally be one and only one entity name,
all subfields may be repeated, appearing as additional fields
sharing identical values in subfields 1 and 2. Where ordering is
of importance (as in sample screen displays), the order of the
table fields (rows) controls precedence. Text which may be
reformatted for display using a variable line length is
represented as an unbroken linear string within a field. Text
which should not be reformatted (for example, the sample screen
displays appearing in SAM fields) assumes that the target system
is capable of displaying at least 80 characters per line. Other
than this, the ISM record contains no notion of text formatting.
Also note that sample screen displays (subfield 2: SAM) have
sometimes been transcribed from printed examples (some of them
produced using variable-width fonts); thus it is impossible to
ascertain if spacing between printing characters arises from
SPACE or TAB characters. These sample displays may also contain
transcription errors.
Leading and trailing spaces have been stripped from all fields in
the file, except for leading spaces appearing in sample screen
displays. Vertical bar characters appearing in sample screen
displays have been changed to colon (:, ASCII 072) characters.
In the case where a particular database, program or file is
available from multiple providers, (e.g., MEDLINE provided by
Dialog, MEDLINE provided by NLM) each entity/provider pair will
have its own ISMR record. This convention is adopted because
providers often use the same name for databases and services
which are actually quite different in scope, coverage, mode of
interaction, or output format.
Specified subfields may contain an indirection operator, a @
character followed by the unique identifier of another logical
ISM record. A proposed convention not currently used in the ISMR
file is to employ a # character followed by the name of an
information source which is available in machine-readable form,
but which is external to the ISM files.
April 22, 1993
5.3.2. ISDEFR (Field Definitions for Main ISM Records)
ISDEFR is the data dictionary defining the universe of legal
subfield names for ISM main records. A core set of subfield
names is provided in this prototype; other organizations may
define additional subfields, where subfield names do not conflict
with pre-defined ones. The NLM will maintain a master copy of
ISM subfield definitions to prevent ambiguous/conflicting use of
subfield names and field formats.
All field identifiers are 3 characters, with the prototype set
intended to be congruent with existing NLM/MEDLARS field names.
Where existing field identifiers are 2 characters, the third
character is an underscore (_, ASCII 137).
The subfield separator for ISDEFR is the caret (^, ASCII 94), and
each row (field) contains the following subfields:
Subfield 1: Three-character subfield identifier.
Subfield 2: Subfield Name.
Subfield 3: Subfield Format.
Subfield 4: Scope information.
Subfield 5: Creator of field definition (initials of
organization, followed by |, followed by initials of
person).
Subfield 6: Valid value field. If T (True), there is a list of
valid values for this field contained in the ISVALR
table. If F (False), the field contents are not a
controlled vocabulary.
5.3.3. ISVALR (Table of Valid Entries for Fields Using Defined
Text)
ISVALR is a lookup table which defines the valid codes which may
appear within subfields which employ a restricted vocabulary.
The subfield separator for ISVALR is the caret (^, ASCII 94), and
each row (field) contains the following subfields:
Subfield 1: Three-character subfield identifier (from set defined
in ISDEFR).
Subfield 2: Valid code.
Subfield 3: Valid text associated with valid code.
Subfield 4: Descriptive commentary, including subfield format
information if appropriate.
April 22, 1993
5.4. List of Current Contents of the ISM
5.4.1. Information Sources, Listed in Order of Unique Identifier
E00001 MEDLINE
E00002 MeSH VOCABULARY
E00005 MED80
E00008 MED66
E00009 AIDSDRUGS
E00011 AIDSTRIALS
E00012 ChemID
E00013 CHEMLINE
E00014 Developmental and Reproductive Toxicology Database
E00016 Environmental Mutagen Information Center Database
Backfile
E00017 Environmental Teratology Information Center Database
Backfile
E00018 GENE-TOX
E00019 HISTLINE
E00020 Hazardous Substances Data Bank
E00021 Integrated Risk Information System
E00022 Name Authority File
E00023 Registry of Toxic Effects of Chemical Substances
E00024 SDILINE
E00025 TOXLINE
E00026 TOXLINE65
E00027 TOXLIT
E00028 TOXLIT65
E00030 AIDSLINE
E00031 AVLINE
April 22, 1993
E00032 BIOETHICS
E00033 CANCERLIT
E00034 CATLINE
E00035 DENTALPROJ
E00036 Directory of Information Resources OnLINE
E00037 HEALTH PLANNING & ADMINISTRATION
E00038 POPLINE
E00039 SERLINE
E00040 Physician Data Query
E00041 BIOTECHSEEK
E00042 DOCUSER
E00043 Chemical Carcinogenesis Research Information System
E00044 Toxic Chemical Release Inventory 1989
E00045 Toxic Chemical Release Inventory 1988
E00046 Toxic Chemical Release Inventory 1987
E00047 Quick Medical Reference (R) Program and Knowledge Base
E00048 Online Mendelian Inheritance in Man
E00049 DXplain
E00050 AI/RHEUM
E00051 Iliad
E00052 GenBank
E00053 Protein Identification Resource
E00054 The EMBL Nucleotide Sequence Database
E00055 DNA Data Bank of Japan
E00056 Toxic Chemical Release Inventory 1990
E00057 TRIFACTS
E00059 Occupational Safety and Health (NIOSH)
April 22, 1993
E00060 Ground Water On-Line
E00061 Oil and Hazardous Materials Technical Assistance Data
System
E00062 National Pesticide Information and Retrieval System
(NPIRS)
E00064 National Environmental Data Referral Service (NEDRES)
E00065 Envirofate
E00066 Computer-Aided Environmental Legislative Data System
E00067 MED85
E00069 MED75
E00070 Toxic Chemical Release Inventory 1991
E00071 Coag-Advisor
E00072 Hepatitis-Advisor
E00073 Technology Transfer Network
5.4.2. Descriptions of Information Suppliers
E00500 National Library of Medicine
E00501 Massachusetts General Hospital Laboratory of Computer
Science
E00502 Chemical Information Systems, Inc.
E00503 U.S. Environmental Protection Agency
E00504 Environmental Technical Information System (ETIS),
Dept. of Urban and Regional Planning, University of
Illinois at Urbana-Champaign
E00505 National Ground Water Information Center (NGWIC),
Ground Water Publishing Company, National Ground Water
Association
E00506 DIALOG Information Services, Inc.
E00507 National Institute for Occupational Safety and Health
(NIOSH)
E00509 BRS Information Technologies, Inc.
E00510 U.S. National Oceanic and Atmospheric Administration
April 22, 1993
E00511 Center for Environmental and Regulatory Information
Systems, Purdue University
E00512 Welch Medical Library, Johns Hopkins University
E00513 Yale Center for Medical Informatics
5.5. Changes from the Last Release
Records were updated where appropriate. A number of new pre-
defined codes were added to the list for the CT_ field, and
several were removed. A Source_of variant was created for the
REN field.
The following ISM entries were removed to create the current
edition:
E00003 MED86
E00004 MED83
E00006 MED77
E00007 MED72
E00058 De Haen's Drug Data
E00063 National Air Toxics Information Clearinghouse Database
E00508 Paul De Haen International, Inc.
The following ISM entries are new:
E00067 MED85
E00069 MED75
E00070 TRI91
E00071 Coag-Advisor
E00072 Hepatitis-Advisor
E00073 Technology Transfer Network (includes the National Air
Toxics Information Clearinghouse Database)
E00513 Yale Center for Medical Informatics
April 22, 1993