Difference between revisions of "Hyeju Jang et al IRI 2006"
PastStudents (talk | contribs) (Created page with '== Citation == Kerstin Denecke and Jochen Bernauer. 2007. Extracting Specific MEdical Data Using Semantic Structures. Artificial Intelligence in Medicine, LNCS Vol. 4594/2007, 2…') |
PastStudents (talk | contribs) |
||
| (11 intermediate revisions by the same user not shown) | |||
| Line 1: | Line 1: | ||
== Citation == | == Citation == | ||
| − | + | Hyeju Jang, Yun Jin, Sung Hyon Myaeng. 2006. Integration of Low Level Linguistic Information for Clinical Document Semantic Tagging. In Proceeding of IEEE Conference on Information Reuse and Integration, 292-297. | |
== Online version == | == Online version == | ||
| − | [http:// | + | [http://ieeexplore.ieee.org/iel5/4018442/4018443/04018505.pdf IEEE Xplore] |
== Summary == | == Summary == | ||
| + | The [[Category::Paper]] presents a semantic tagger which extract "symptom", "therapy", and "performance" from free-text clinical records. It can be said high-level [[AddressesProblem::Named Entity Recognition]] based on phrase in order to help answering two questions below. | ||
| − | + | 1. How can X be used in the treatment of Y? | |
| − | + | 2. What are the performance characteristics of X in the setting of Y? | |
| − | |||
| − | + | The data set used here was 300 narrative sections of "progress after hospital stay" of Clinical Data Architecture (CDA) documents, which came from Seoul National University Hospital. The data is not public. | |
| + | Texts were written by Korean doctors. The characteristics of the texts are they have a lot of specialized medical words, abbreviations, and non-alphanumeric symbols. In addition, the texts are written in mixed Korean and English. Mostly, English are used for the medical terminologies, and Korean are for some general nouns and most verbs though there are some exceptions depending on the style of a doctor. Some examples are below. | ||
| − | The | + | * Specialized medical words |
| + | **Ex) hypothyroidism , hypertensive | ||
| + | * Abbreviations | ||
| + | ** Ex) ACA, MRI, CR | ||
| + | * non-alphanumeric symbols | ||
| + | ** Ex) , ↑ | ||
| + | * numeric data | ||
| + | ** Ex) 0.24/73.2 , 1-2회 , 12.1% , 02.12.16-03.1.2 | ||
| + | * Mixed Korean and English | ||
| + | ** English for the medical terminologies | ||
| + | * Korean for some general nouns and most verbs | ||
| + | ** Ex) 시행한 basal hormone level 에서, clinical 하게는 cushinoid feature 처럼 보이나 | ||
| + | |||
| + | The approach of the system is [[UsesMethod::Hidden Markov Model]] (HMM) using equivalence classes based on primitive tagging results such as UMLS and POS to solve serious data sparse problem. | ||
| + | |||
| + | [[File:seta.png]] | ||
| + | |||
| + | The evaluation was performed on 200 documents for training and 100 documents for test with 3 fold validation. The performance of the system is not high, approximately 70%. | ||
| + | The author thinks this could be improved with reflecting various aspects of language. | ||
Latest revision as of 16:06, 21 October 2010
Citation
Hyeju Jang, Yun Jin, Sung Hyon Myaeng. 2006. Integration of Low Level Linguistic Information for Clinical Document Semantic Tagging. In Proceeding of IEEE Conference on Information Reuse and Integration, 292-297.
Online version
Summary
The Paper presents a semantic tagger which extract "symptom", "therapy", and "performance" from free-text clinical records. It can be said high-level Named Entity Recognition based on phrase in order to help answering two questions below.
1. How can X be used in the treatment of Y?
2. What are the performance characteristics of X in the setting of Y?
The data set used here was 300 narrative sections of "progress after hospital stay" of Clinical Data Architecture (CDA) documents, which came from Seoul National University Hospital. The data is not public.
Texts were written by Korean doctors. The characteristics of the texts are they have a lot of specialized medical words, abbreviations, and non-alphanumeric symbols. In addition, the texts are written in mixed Korean and English. Mostly, English are used for the medical terminologies, and Korean are for some general nouns and most verbs though there are some exceptions depending on the style of a doctor. Some examples are below.
- Specialized medical words
- Ex) hypothyroidism , hypertensive
- Abbreviations
- Ex) ACA, MRI, CR
- non-alphanumeric symbols
- Ex) , ↑
- numeric data
- Ex) 0.24/73.2 , 1-2회 , 12.1% , 02.12.16-03.1.2
- Mixed Korean and English
- English for the medical terminologies
- Korean for some general nouns and most verbs
- Ex) 시행한 basal hormone level 에서, clinical 하게는 cushinoid feature 처럼 보이나
The approach of the system is Hidden Markov Model (HMM) using equivalence classes based on primitive tagging results such as UMLS and POS to solve serious data sparse problem.
The evaluation was performed on 200 documents for training and 100 documents for test with 3 fold validation. The performance of the system is not high, approximately 70%. The author thinks this could be improved with reflecting various aspects of language.
