Data-Driven Information Extraction from Chinese Electronic Medical Records
OBJECTIVE: This study aims to propose a data-driven framework that takes unstructured free text narratives in Chinese Electronic Medical Records (EMRs) as input and converts them into structured time-event-description triples, where the description is either an elaboration or an outcome of the medical event.
MATERIALS AND METHODS: Our framework uses a hybrid approach. It consists of constructing cross-domain core medical lexica, an unsupervised, iterative algorithm to accrue more accurate terms into the lexica, rules to address Chinese writing conventions and temporal descriptors, and a Support Vector Machine (SVM) algorithm that innovatively utilizes Normalized Google Distance (NGD) to estimate the correlation between medical events and their descriptions.
RESULTS: The effectiveness of the framework was demonstrated with a dataset of 24,817 de-identified Chinese EMRs. The cross-domain medical lexica were capable of recognizing terms with an F1-score of 0.896. 98.5% of recorded medical events were linked to temporal descriptors. The NGD SVM description-event matching achieved an F1-score of 0.874. The end-to-end time-event-description extraction of our framework achieved an F1-score of 0.846.
DISCUSSION: In terms of named entity recognition, the proposed framework outperforms state-of-the-art supervised learning algorithms (F1-score: 0.896 vs. 0.886). In event-description association, the NGD SVM is superior to SVM using only local context and semantic features (F1-score: 0.874 vs. 0.838).
CONCLUSIONS: The framework is data-driven, weakly supervised, and robust against the variations and noises that tend to occur in a large corpus. It addresses Chinese medical writing conventions and variations in writing styles through patterns used for discovering new terms and rules for updating the lexica.
Medienart: |
E-Artikel |
---|
Erscheinungsjahr: |
2015 |
---|---|
Erschienen: |
2015 |
Enthalten in: |
Zur Gesamtaufnahme - volume:10 |
---|---|
Enthalten in: |
PloS one - 10(2015), 8 vom: 16., Seite e0136270 |
Sprache: |
Englisch |
---|
Beteiligte Personen: |
Xu, Dong [VerfasserIn] |
---|
Links: |
---|
Themen: |
---|
Anmerkungen: |
Date Completed 13.05.2016 Date Revised 13.11.2018 published: Electronic-eCollection Citation Status MEDLINE |
---|
doi: |
10.1371/journal.pone.0136270 |
---|
funding: |
|
---|---|
Förderinstitution / Projekttitel: |
|
PPN (Katalog-ID): |
NLM25204763X |
---|
LEADER | 01000naa a22002652 4500 | ||
---|---|---|---|
001 | NLM25204763X | ||
003 | DE-627 | ||
005 | 20231224163311.0 | ||
007 | cr uuu---uuuuu | ||
008 | 231224s2015 xx |||||o 00| ||eng c | ||
024 | 7 | |a 10.1371/journal.pone.0136270 |2 doi | |
028 | 5 | 2 | |a pubmed24n0840.xml |
035 | |a (DE-627)NLM25204763X | ||
035 | |a (NLM)26295801 | ||
040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
041 | |a eng | ||
100 | 1 | |a Xu, Dong |e verfasserin |4 aut | |
245 | 1 | 0 | |a Data-Driven Information Extraction from Chinese Electronic Medical Records |
264 | 1 | |c 2015 | |
336 | |a Text |b txt |2 rdacontent | ||
337 | |a ƒaComputermedien |b c |2 rdamedia | ||
338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
500 | |a Date Completed 13.05.2016 | ||
500 | |a Date Revised 13.11.2018 | ||
500 | |a published: Electronic-eCollection | ||
500 | |a Citation Status MEDLINE | ||
520 | |a OBJECTIVE: This study aims to propose a data-driven framework that takes unstructured free text narratives in Chinese Electronic Medical Records (EMRs) as input and converts them into structured time-event-description triples, where the description is either an elaboration or an outcome of the medical event | ||
520 | |a MATERIALS AND METHODS: Our framework uses a hybrid approach. It consists of constructing cross-domain core medical lexica, an unsupervised, iterative algorithm to accrue more accurate terms into the lexica, rules to address Chinese writing conventions and temporal descriptors, and a Support Vector Machine (SVM) algorithm that innovatively utilizes Normalized Google Distance (NGD) to estimate the correlation between medical events and their descriptions | ||
520 | |a RESULTS: The effectiveness of the framework was demonstrated with a dataset of 24,817 de-identified Chinese EMRs. The cross-domain medical lexica were capable of recognizing terms with an F1-score of 0.896. 98.5% of recorded medical events were linked to temporal descriptors. The NGD SVM description-event matching achieved an F1-score of 0.874. The end-to-end time-event-description extraction of our framework achieved an F1-score of 0.846 | ||
520 | |a DISCUSSION: In terms of named entity recognition, the proposed framework outperforms state-of-the-art supervised learning algorithms (F1-score: 0.896 vs. 0.886). In event-description association, the NGD SVM is superior to SVM using only local context and semantic features (F1-score: 0.874 vs. 0.838) | ||
520 | |a CONCLUSIONS: The framework is data-driven, weakly supervised, and robust against the variations and noises that tend to occur in a large corpus. It addresses Chinese medical writing conventions and variations in writing styles through patterns used for discovering new terms and rules for updating the lexica | ||
650 | 4 | |a Journal Article | |
700 | 1 | |a Zhang, Meizhuo |e verfasserin |4 aut | |
700 | 1 | |a Zhao, Tianwan |e verfasserin |4 aut | |
700 | 1 | |a Ge, Chen |e verfasserin |4 aut | |
700 | 1 | |a Gao, Weiguo |e verfasserin |4 aut | |
700 | 1 | |a Wei, Jia |e verfasserin |4 aut | |
700 | 1 | |a Zhu, Kenny Q |e verfasserin |4 aut | |
773 | 0 | 8 | |i Enthalten in |t PloS one |d 2006 |g 10(2015), 8 vom: 16., Seite e0136270 |w (DE-627)NLM167327399 |x 1932-6203 |7 nnns |
773 | 1 | 8 | |g volume:10 |g year:2015 |g number:8 |g day:16 |g pages:e0136270 |
856 | 4 | 0 | |u http://dx.doi.org/10.1371/journal.pone.0136270 |3 Volltext |
912 | |a GBV_USEFLAG_A | ||
912 | |a GBV_NLM | ||
951 | |a AR | ||
952 | |d 10 |j 2015 |e 8 |b 16 |h e0136270 |