1
Generally, every hospital generates a clinical document like discharge summaries and progress notes for every patient. The maintenance of the clinical documents is a tough task for the doctors to manage. There are also chances that the documents can be lost, so in order to manage some issues the doctors had been using the Health Level 7 (HL7) standards since few years. Health Level Seven International (HL7) is an ANSI-accredited standards developing organization (SDO) that creates health care messaging standards. The doctors now are using the HL7 version2, where in version2 there’s no consistency and version2 is mostly based on the framework and has a less interface to interface structure. Over a decade, the work has been going on the next version of the HL7 which is HL7 version3. HL7 version3 uses the Clinical Document Architecture (CDA) .The HL7 Clinical Document Architecture (CDA) is a XML based mark-up standard that specifies the structure and semantics of clinical documents for exchange Introduction Objectives Design References Data sets: https://i2b2.org/NLP/DataSets/ Clinical Codes: Rx-Norm- http://bioportal.bioontology.org/ontologies/ Snomed CT- http://vtsl.vetmed.vt.edu/ Loinc- http://search.loinc.org/ https://www.cori.org/excellencereport/old/DocumentArchitecture_BDo lin.pdf Acknowledgements The success and final outcome of this project required a lot of guidance and assistance. We would like to thank Dr. Praveen Rao for offering such an interesting course(Health Informatics) .Moreover we would like to thank him for the support and guidance through out the semester. Dr. Praveen Rao- http://r.web.umkc.edu/raopr We would like to thank the i2b2(A national center for Biomedical Computing) for providing the research data sets. School of Computing and Engineering, University of Missouri - Kansas City Nikhil Kassetty, Vamsi Vura, Sandeep Keshetti, Rakesh Gandu HL7 CDA GEN The main objective of this project is to generate the synthetic documents in CDA structure. We also developed interfaces where the user can provide parameters for generating the synthetic documents. Generate valid and meaningful XML documents. The scope of the project is to generate CDA documents , the generated documents should have the characteristics like context oriented, wholeness and human readable. Clinicians should have good knowledge on the standard clinical coding systems like SNOMED_CT , LOINC , ICD-9 and Rx-NORM. The synthetic generator will be able to generate one million documents for the diseases Diabetes, Asthma and Cancer. Motivation Health has a great significance in everyone's life, Good health have lot of benefits and it is very important aspect in the life span of a living being . In order to provide good health , it’s important to promote good ideas and technologies on health. This motivation helps us to think much and more on the wellness of the life and it encourages to develop and update with the new technologies. Here, we developed a synthetic data generator which provides a framework for the incremental growth in the amount and precision of structured , vocabulary bound clinical information exchange. We collected the synthetic datasets where the data consists of : Discharge summary. Clinical summary report. History and physical examination. Diagnostic Reports. Medications. Data representation. Composition Of Clinical Document Architecture(CDA) Level 1: Root of the hierarchy and is the most general architecture. Level one supports full CDA semantics including document sections and structured entries. Level 2: Provides additional constraints on a document by creating distinct templates at the section level for each type of the document. Level 3: Provides additional constraints at the entry level and the optional constraints at the section level. In the CDA document, we use the clinical coding systems like: SNOMED CTSystematized Nomenclature of Medicine Clinical Terms . SNOMED covers diseases, clinical findings, procedures, drugs, etc. LOINCLogical Observations: Identifiers, Names and Codes. LOINC is divided into lab, and clinical portions. RX-NORMRx Norm is a name of US-specified terminology in medicine that contains all medications available in US market. ICD-9International Classification Of Diseases 9 th revision. ICD9 is used for the international classification of diseases. In our project, CDA document consists of the following sections: Admission Date, Principal Diagnosis , History of Present Illness, Past History, Past Surgical History, Medications on Admission, Physical Examination, Hospital Course, Medications on Discharge, Discharge Date. JAVA technologies: Servlets, JSP. Web technologies: HTML, JavaScript and CSS. Servers: Tomcat User Interface of the Synthetic Data Generator(CDA GEN) Results When the user gives his requirements through the user interface and click generate button the generated documents are stored in a specific folder . The files generated are in the XML format and the files can be viewed with the browser or with the XML editors. The XML file generated will contain all the sections and it’s text along with clinical codes. Future Work We have used very limited datasets , it would be more effective if the datasets are large in number. We have used the text files to store the data, it would be easier if the datasets are stored in the database. We stored the generated documents in the local drive, it would be easier to maintain and organize if the documents are stored in the database. We have worked on level 1 and level2 CDA documents, should work with level 3. Usefulness Clinical Document Architecture forms the basis for the Continuity of Care Document standard for patient document information exchange. To patients : The CDA architecture can be useful to the patients where there is no need for the patients to carry the health records along with them because with the help of CDA every patient has unique ID so that he can just use that ID to know his health record and history. To doctors: It’s useful to doctors such that the doctors need not store all the patient records in the form of paper and when a patient approaches a doctor for the first time then the doctor can see all the health record and history of the patient with the help of the CDA documents and the clinical coding standards and also the CDA documents can be exchanged between different systems with the help of unique IDs. Technologies Used

HL7 Synthetic CDA generator -Final presentation

Embed Size (px)

Citation preview

Page 1: HL7 Synthetic CDA generator -Final presentation

• Generally, every hospital generates a clinical document like discharge

summaries and progress notes for every patient. The maintenance of the

clinical documents is a tough task for the doctors to manage. There are

also chances that the documents can be lost, so in order to manage some

issues the doctors had been using the Health Level 7 (HL7) standards

since few years.

• Health Level Seven International (HL7) is an ANSI-accredited standards

developing organization (SDO) that creates health care messaging

standards. The doctors now are using the HL7 version2, where in

version2 there’s no consistency and version2 is mostly based on the

framework and has a less interface to interface structure.

• Over a decade, the work has been going on the next version of the HL7

which is HL7 version3. HL7 version3 uses the Clinical Document

Architecture (CDA) .The HL7 Clinical Document Architecture (CDA)

is a XML based mark-up standard that specifies the structure and

semantics of clinical documents for exchange

Introduction

Objectives

Design

References

Data sets: https://i2b2.org/NLP/DataSets/

Clinical Codes:

Rx-Norm- http://bioportal.bioontology.org/ontologies/

Snomed CT- http://vtsl.vetmed.vt.edu/

Loinc- http://search.loinc.org/

https://www.cori.org/excellencereport/old/DocumentArchitecture_BDo

lin.pdf

Acknowledgements

The success and final outcome of this project required a lot of guidance

and assistance.

We would like to thank Dr. Praveen Rao for offering such an

interesting course(Health Informatics) .Moreover we would like to thank

him for the support and guidance through out the semester.

Dr. Praveen Rao- http://r.web.umkc.edu/raopr

We would like to thank the i2b2(A national center for

Biomedical Computing) for providing the research data sets.

School of Computing and Engineering, University of Missouri - Kansas City

Nikhil Kassetty, Vamsi Vura, Sandeep Keshetti, Rakesh Gandu

HL7 CDA GEN

•The main objective of this project is to generate the synthetic documents

in CDA structure. We also developed interfaces where the user can provide

parameters for generating the synthetic documents.

•Generate valid and meaningful XML documents.

•The scope of the project is to generate CDA documents , the generated

documents should have the characteristics like context oriented, wholeness

and human readable.

•Clinicians should have good knowledge on the standard clinical coding

systems like SNOMED_CT , LOINC , ICD-9 and Rx-NORM.

•The synthetic generator will be able to generate one million documents

for the diseases Diabetes, Asthma and Cancer.

Motivation

• Health has a great significance in everyone's life, Good health have lot of

benefits and it is very important aspect in the life span of a living being .

• In order to provide good health , it’s important to promote good ideas and

technologies on health. This motivation helps us to think much and more

on the wellness of the life and it encourages to develop and update with the

new technologies.

• Here, we developed a synthetic data generator which provides a

framework for the incremental growth in the amount and precision of

structured , vocabulary bound clinical information exchange.

We collected the synthetic datasets where the data consists of :

•Discharge summary.

•Clinical summary report.

•History and physical examination.

•Diagnostic Reports.

•Medications.

Data representation.

Composition Of Clinical Document Architecture(CDA)

Level 1:

Root of the hierarchy and is the most general architecture. Level one

supports full CDA semantics including document sections and structured

entries.

Level 2:

Provides additional constraints on a document by creating distinct

templates at the section level for each type of the document.

Level 3:

Provides additional constraints at the entry level and the optional

constraints at the section level.

In the CDA document, we use the clinical coding systems like:

•SNOMED CT– Systematized Nomenclature of Medicine Clinical

Terms .

SNOMED covers diseases, clinical findings, procedures, drugs, etc.

•LOINC–Logical Observations: Identifiers, Names and Codes.

LOINC is divided into lab, and clinical portions.

•RX-NORM–Rx Norm is a name of US-specified terminology

in medicine that contains all medications available in US

market.

ICD-9–International Classification Of Diseases 9th revision.

ICD9 is used for the international classification of diseases.

In our project, CDA document consists of the following sections:

Admission Date, Principal Diagnosis , History of Present Illness,

Past History, Past Surgical History, Medications on Admission,

Physical Examination, Hospital Course, Medications on Discharge,

Discharge Date.

JAVA technologies: Servlets, JSP.

Web technologies: HTML, JavaScript and CSS.

Servers: Tomcat

User Interface of the Synthetic Data Generator(CDA GEN)

Results

When the user gives his requirements through the user interface and click

generate button the generated documents are stored in a specific folder .

The files generated are in the XML format and the files can be viewed with

the browser or with the XML editors.

The XML file generated will contain all the sections and it’s text along

with clinical codes.

Future Work

•We have used very limited datasets , it would be more effective if the

datasets are large in number.

•We have used the text files to store the data, it would be easier if the

datasets are stored in the database.

•We stored the generated documents in the local drive, it would be easier to

maintain and organize if the documents are stored in the database.

•We have worked on level 1 and level2 CDA documents, should work with

level 3.

Usefulness

Clinical Document Architecture forms the basis for the Continuity of

Care Document standard for patient document information exchange.

To patients : The CDA architecture can be useful to the patients where

there is no need for the patients to carry the health records along with them

because with the help of CDA every patient has unique ID so that he can

just use that ID to know his health record and history.

To doctors: It’s useful to doctors such that the doctors need not store all

the patient records in the form of paper and when a patient approaches a

doctor for the first time then the doctor can see all the health record and

history of the patient with the help of the CDA documents and the clinical

coding standards and also the CDA documents can be exchanged between

different systems with the help of unique IDs.

Technologies Used