EXPERIMENTAL DATABASE MEDICAL SYSTEM
Data Acquisition Background and Features
Petr Maule, Jiri Polivka
Department of Computer Science and Eng., University of West Bohemia, Univerzitni 8, Plzen, Czech Republic
Jana Kleckova
Department of Computer Science and Eng., University of West Bohemia, Univerzitni 8, Plzen, Czech Republic
Keywords: Medical Software, Medical Standards, DASTA, DICOM, HL7.
Abstract: This article describes features and background of the upcoming scientific medical system. The ambitious
goal of the system is to provide relevant medical data for various medical methods and experimental
implementations testing. The article presents needs for such system and administrative background which it
requires. The main standards which are used in medical applications are also mentioned, like the DICOM,
DASTA or HL7 format with the relations to our experimental medical system.
1 OVERVIEW
Medical science is quickly evolving branch
providing new treatment techniques and new
diagnosis determination possibilities. Progress of
software computation methods and evolution of new
software algorithms plays the major role. All kinds
of new techniques need to be tested and verified at
large amount of suitable data. These data differs in
context of methods to be verified. Some methods
can strictly require one kind of input data – like one
type of examination as computed tomography (CT)
or x-ray images, etc. Other methods require analysis
of several examination types together. Some
methods can even require sequences of the same
kind of examinations acquired in different time or
they can even have more sophisticated needs for
input data.
The goal of our ongoing medical system is to
offer enough relevant data for all kind of input data
requirements. If there is any new method which
needs to be tested on larger amount of data, the
system can provide it. It will store anonymous data
coming from all our partners – mostly hospitals. The
system will be sophisticated in processing of
heterogeneous multimedia medical data objects.
Also a knowledge-based system using the database
will be proposed. Following parts describe existing
examination types and storing formats, the way how
the data will be transferred from our partners into the
system and description of benefits for all system
users.
2 MEDICAL DATA TYPES AND
STANDARDS
Hospital information systems at the time of their
first introduction had incompatible data formats. The
same problem was with medical instruments – no
mutually compatible formats. Recently the situation
is much better because many standards have been
established. The leading standard in digital imaging
is DICOM standard which provides the same format
for many imaging modalities. DICOM standard is
supported in vast majority of hospital information
systems and medical instruments worldwide. But
there are still data which are non-imaging and need
to be stored as well. Example of such data can be
laboratory examinations or medical diagnosis text
records, etc. For these kinds of data there are mostly
used standards like HL7 and especially in the Czech
Republic DASTA standard. Our system is going to
support DICOM standard and DASTA standard with
future plan for HL7 standard implementation.
402
Maule P., Polivka J. and Kleckova J. (2010).
EXPERIMENTAL DATABASE MEDICAL SYSTEM - Data Acquisition Background and Features.
In Proceedings of the Third International Conference on Health Informatics, pages 402-405
DOI: 10.5220/0002698004020405
Copyright
c
SciTePress
2.1 DICOM Standard
The Digital Imaging and Communications in
Medicine (DICOM) standard begins in 1983 when
The American College of Radiology (ACR) and
National Electrical Manufacturers Association
(NEMA) formed a committee to develop standard
unifying digital information interoperability. Current
version of this standard is DICOM 3.0.
The standard covers many parts connected with
medical imaging information interoperability. It
defines context of information objects for different
modalities as well as possible services operating
with these data objects and it also defines
conformance statement which must be claimed by
DICOM compatible instruments. It also defines
communication interface and media storing
specifications.
2.2 DASTA Standard
The data standard DASTA (DS), established by the
Ministry of Health of the Czech Republic, is used
for transferring important data among various
medical information systems in the university
hospitals as well as in smaller medical facilities. The
current version of the format is DS4 (DS 04.02.04)
and the development is still in progress. The DS
format simplifies communication between many
medical information systems used in the Czech
Republic, which was very complicated before (one
of the medical information systems had more than
30 different protocols and diagnoses code-lists).
The first simple version of DS standard (DS 1.0)
was published in the bulletin of the Ministry of
Health of the CR in 1994. DS 1.0 set up the
conceptual principia of the system, but it was still
weak in data blocks and it didn´t solve the
problematic of the laboratory segment. The next
development tended to set up and improve complex
communication interface with the medical laboratory
information systems and to creation of the national
laboratory items code-list (NLIC). The development
of the first practically usable version of DS (DS1.1.)
and NLIC was completed in 1997. The early version
of DS standard came out from basic text files in
TXT, the next versions used the XML data format
(DS 2 and 3 used DTD, DS 4 are fully exploiting
XML). Important versions, their launch dates and
data formats are:
DS 01.10 1. 7. 1997 "TXT"
DS 01.20 1. 1. 2001 "TXT"
DS 02.01 1. 5. 2002 "DTD"
DS 03.01 2. 6. 2003 "DTD"
DS 04.01 1. 1. 2007 "XML"
The next important moment for expansion and
quality enhancement of DS standard was the
supportive web-services introduction. These services
offer full and actual DS4, available documentation,
detailed manual and implemental programs. The
regular four times a year updates are realized by the
web-services too.
At the present time, DASTA standard contains
not only blocks for patient data transfer, but also
blocks for investigation of the drinking and supply
water, diagnoses and laboratory code-lists and
others. In the patient data scope, the current DS
standard allows transfer of the important information
from many areas. For example the patient
identification data, urgent information (allergy,
diagnosis), health insurance company information,
anamnesis, patient medicaments, persistent and
acute diagnosis, vaccinations or special data blocks.
As the example of clinical events, the DS standard
could be used for the laboratory work-up, RDG
examination (RTG, CT and other), pulmonary
function tests, consultation, ambulant and dismissory
report and other.
The main advantage of the DS standard in the
Czech Republic is its domestic background. On its
development participated many medical software
developers, so the DS meet their particular
requirements. Some medical software which use the
DS standard are DATAPLAN, HICOMP, ICZ,
LOGIS, MEDICALC, MEDICON, PCS, SOPHIS,
DS SOFT, DYNATECH and others. DS standard is
used also in Slovakia, although it is not the formal
standard there. The next step could be distribution of
the DS to other counties in the EU.
2.3 HL7 Standard
The HL7 is another international standard in the
clinical and administrative medical data domain. It
was developed by Health Level Seven Inc.
organization, which is accredited by ANSI
(American National Standards Institute) as a SDO
(Standards Developing Organization). This Health
Level Seven Inc. was founded in 1987 as a not-for-
profit volunteer organization like other ANSI-
accredited SDOs. The standard are developed in co-
operation with many various participants, who may
include providers, consultants, vendors, payers and
government groups who are interested in clinical and
administrative standards for health care in the USA
and word wide.
As the other standards, HL7 was developed like
messaging protocol that enable to share and
exchange sets of clinical and administrative data
EXPERIMENTAL DATABASE MEDICAL SYSTEM - Data Acquisition Background and Features
403
among unrelated health care applications and
facilities. The members of HL7 are known as the
Working Group and are organized into special
interest groups (SIGs) and technical committees.
They are looking at end-user needs and hand it on to
the attention of the HL7 Working Group. The HL7
standard is a text-based encoding system.
The first version of the HL7 standard (V.2.x) was
developed and approved as an ANSI standard more
than 12 years ago. The current version (V.3.0) uses
an object-oriented development methodology and a
Reference Information Model (RIM) to create
messages. The definition of the HL7 messages was
relatively free in version V.2.x whereas version
V.3.0 reduces optionality of the message content
which means fewer errors in the messages
interpretation. In the current versions the XML
protocol is used to envelope HL7 message content
which increase system efficiency.
Because of its text-based format, the HL7 is still
human-readable, although it is usually interpreted by
machines and medical software. Several web sites
provide testing HL7 messages and can be used to
test a system´s ability to process the HL7 standard.
3 THE SYSTEM FILLING
As it was already said the system will be fed by our
partners which will be mostly hospitals. Now at the
beginning of the system implementation we have
only one partner – The University Hospital in Pilsen.
For all the partners we need to implement special
export module to respect their information system.
The University Hospital in Pilsen uses WinMedicalc
(WM) information system. WM is used at many
hospitals in Czech Republic. The export module can
be used at all the hospitals which use WM and
decide to become our partner with no need for
modifications. Generally we can say that all
information system mostly covers DICOM, DASTA
and HL7 standards. The export module just exports
data of these standards into our system with
preservation of mutual relation but in anonymous
form.
There are two requirements of our system on
filling process. The first is that data must be able to
be exported on demand and the second that all
targeted data are exported automatically. There must
be also duplicity removal in export module. Export
module also ensures transferring from partner´s
environment into our system environment which
covers some kind of on-line transfers. DICOM and
DASTA standard data are processed separately. The
filling schema of the system is shown in the figure 1.
Figure 1: Filling system schema.
3.1 DASTA Exporting
The DASTA format for export is common XML
format. WM provides export services which
repeatedly look for new DASTA standard data in
hospital information system and all found records it
exports using web services or directly stores it into
preselected directory at selected computer in
common XML file form. At this computer our
export module is running. Export module just waits
for new DASTA records, process them and transfers
it into our system. WM exporting ensures no
duplicity of DASTA information.
3.2 DICOM Exporting
The Export module for DICOM data is more
general. It can be used everywhere where DICOM
standard is used. It is not depended on used hospital
information system. Export module contains
DICOM Application Entity (AET) which is able to
communicate via DICOM standard communication
interface. This module repeatedly asks DICOM
image server for new records, download it, process
and transfers it into our system.
Here is one feature that needs to be solved.
DICOM searching for new data does not ensure
unique data. DICOM communication provides query
system just for data which were acquired at selected
date but the data could be exported also on demand.
There must be stored information in export module
about already exported DICOM data to refuse all
duplicities.
3.3 Patient Identification
The system preserves examinations relations. It
means that we know which examination belongs to
which patient in anonymizated form. During
anonymization process in export module, there is
HEALTHINF 2010 - International Conference on Health Informatics
404
assigned unique signature for each patient. This
mapping will be stored only in export module and
stays inside our partner's system (in hospital).
Assigned signature is randomly generated signature
unique in our system. For all further data of this
patient, there will be found mapping in the export
module. It will use the same signature which means
relations preservation.
3.4 Data Size
We are going to transfer large volume of data.
Different examination means different data sizes.
Probably the highest data sizes will come from
DICOM file where one examination can even have
about 4 GB. These data are to be transferred on-line,
so there is required high-speed connection of all
partners.
3.5 Drawbacks
If the patient undergo two examinations in different
hospital information system scope (often means two
different hospitals), it will be presented as two
different patients – now we do not have enough
information for mapping these like occurrences.
4 CONCLUSIONS
Our system is tending to contain large amount of
different medical data provided in DICOM and
DASTA standards and in the future also HL7
standard. All data will be anonymous but mutual
relations will be preserved. This article shortly
describes some of the examination types and storing
formats which we will take into account in our
scientific database medical system, like the DASTA
and HL7 standard, new DICOM standard is also
noticed. We introduced the approach how the data
could be transferred from medical facilities and
other filling partners into our system and mentioned
the benefits for all system users. We will use these
findings in our future research work.
REFERENCES
The DICOM Standard, http://dicom.nema.org
Vcelak P, Polivka J, Maule P, Kratochvil P and Kleckova
J (2009). Experimental database system for the
vascular brain diseases research. Frontiers in
Neuroinformatics. Conference Abstract: 2nd INCF
Congress of Neuroinformatics. doi:
10.3389/conf.neuro.11.2009.08.002
Data Standard MZČR,
http://ciselniky.dasta.mzcr.czHealth Level 7,
http://www.hl7.org
EXPERIMENTAL DATABASE MEDICAL SYSTEM - Data Acquisition Background and Features
405