Neural transfer learning for assigning diagnosis codes to EMRs

•Transfer learning using convolutional neural networks improves multi-label learning.•Predicting MeSH terms for biomedical articles is a useful source task for EMR coding.•Using 2 copies of source task parameters, one fixed and one tuned, helps target models.•Using both word embeddings and convoluti...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Artificial intelligence in medicine Jg. 96; S. 116 - 122
Hauptverfasser: Rios, Anthony, Kavuluru, Ramakanth
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Netherlands Elsevier B.V 01.05.2019
Schlagworte:
ISSN:0933-3657, 1873-2860, 1873-2860
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract •Transfer learning using convolutional neural networks improves multi-label learning.•Predicting MeSH terms for biomedical articles is a useful source task for EMR coding.•Using 2 copies of source task parameters, one fixed and one tuned, helps target models.•Using both word embeddings and convolutions from source task improves prediction. Electronic medical records (EMRs) are manually annotated by healthcare professionals and specialized medical coders with a standardized set of alphanumeric diagnosis and procedure codes, specifically from the International Classification of Diseases (ICD). Annotating EMRs with ICD codes is important for medical billing and downstream epidemiological studies. However, manually annotating EMRs is both time-consuming and error prone. In this paper, we explore the use of convolutional neural networks (CNNs) for automatic ICD coding. Because many codes occur infrequently, CNN performance is inhibited. Therefore, we propose supplementing EMR data with PubMed indexed biomedical research abstracts through neural transfer learning. Transfer learning is the process of “transferring” knowledge acquired from one task (the source task) to a different (target) task. For the source task, we train a CNN to predict medical subject headings (MeSH) using 1.6 million PubMed indexed biomedical abstracts. For the target task, we train a CNN on 71,463 real-world EMRs collected from the University of Kentucky (UKY) medical center to predict ICD diagnosis codes. We introduce a simple, yet effective, transfer learning methodology which avoids forgetting knowledge gained from the source task. Compared to our prior work using EMRs from the UKY medical center, we improve both the micro and macro F-scores by more than 8%. Likewise, compared to other transfer learning methods, our approach results in nearly 2% improvement in macro F-score. We show that transfer learning can improve CNN performance for EMR coding in the presence of data sparsity issues. Furthermore, we find that our proposed transfer learning approach outperforms other methods with respect to macro F-score. Finally, we analyze how transfer learning impacts codes with respect to code frequency. We find that we achieve greater improvement on infrequent codes compared to improvements in most frequent codes.
AbstractList •Transfer learning using convolutional neural networks improves multi-label learning.•Predicting MeSH terms for biomedical articles is a useful source task for EMR coding.•Using 2 copies of source task parameters, one fixed and one tuned, helps target models.•Using both word embeddings and convolutions from source task improves prediction. Electronic medical records (EMRs) are manually annotated by healthcare professionals and specialized medical coders with a standardized set of alphanumeric diagnosis and procedure codes, specifically from the International Classification of Diseases (ICD). Annotating EMRs with ICD codes is important for medical billing and downstream epidemiological studies. However, manually annotating EMRs is both time-consuming and error prone. In this paper, we explore the use of convolutional neural networks (CNNs) for automatic ICD coding. Because many codes occur infrequently, CNN performance is inhibited. Therefore, we propose supplementing EMR data with PubMed indexed biomedical research abstracts through neural transfer learning. Transfer learning is the process of “transferring” knowledge acquired from one task (the source task) to a different (target) task. For the source task, we train a CNN to predict medical subject headings (MeSH) using 1.6 million PubMed indexed biomedical abstracts. For the target task, we train a CNN on 71,463 real-world EMRs collected from the University of Kentucky (UKY) medical center to predict ICD diagnosis codes. We introduce a simple, yet effective, transfer learning methodology which avoids forgetting knowledge gained from the source task. Compared to our prior work using EMRs from the UKY medical center, we improve both the micro and macro F-scores by more than 8%. Likewise, compared to other transfer learning methods, our approach results in nearly 2% improvement in macro F-score. We show that transfer learning can improve CNN performance for EMR coding in the presence of data sparsity issues. Furthermore, we find that our proposed transfer learning approach outperforms other methods with respect to macro F-score. Finally, we analyze how transfer learning impacts codes with respect to code frequency. We find that we achieve greater improvement on infrequent codes compared to improvements in most frequent codes.
Electronic medical records (EMRs) are manually annotated by healthcare professionals and specialized medical coders with a standardized set of alphanumeric diagnosis and procedure codes, specifically from the International Classification of Diseases (ICD). Annotating EMRs with ICD codes is important for medical billing and downstream epidemiological studies. However, manually annotating EMRs is both time-consuming and error prone. In this paper, we explore the use of convolutional neural networks (CNNs) for automatic ICD coding. Because many codes occur infrequently, CNN performance is inhibited. Therefore, we propose supplementing EMR data with PubMed indexed biomedical research abstracts through neural transfer learning.OBJECTIVEElectronic medical records (EMRs) are manually annotated by healthcare professionals and specialized medical coders with a standardized set of alphanumeric diagnosis and procedure codes, specifically from the International Classification of Diseases (ICD). Annotating EMRs with ICD codes is important for medical billing and downstream epidemiological studies. However, manually annotating EMRs is both time-consuming and error prone. In this paper, we explore the use of convolutional neural networks (CNNs) for automatic ICD coding. Because many codes occur infrequently, CNN performance is inhibited. Therefore, we propose supplementing EMR data with PubMed indexed biomedical research abstracts through neural transfer learning.Transfer learning is the process of "transferring" knowledge acquired from one task (the source task) to a different (target) task. For the source task, we train a CNN to predict medical subject headings (MeSH) using 1.6 million PubMed indexed biomedical abstracts. For the target task, we train a CNN on 71,463 real-world EMRs collected from the University of Kentucky (UKY) medical center to predict ICD diagnosis codes. We introduce a simple, yet effective, transfer learning methodology which avoids forgetting knowledge gained from the source task.MATERIALS AND METHODSTransfer learning is the process of "transferring" knowledge acquired from one task (the source task) to a different (target) task. For the source task, we train a CNN to predict medical subject headings (MeSH) using 1.6 million PubMed indexed biomedical abstracts. For the target task, we train a CNN on 71,463 real-world EMRs collected from the University of Kentucky (UKY) medical center to predict ICD diagnosis codes. We introduce a simple, yet effective, transfer learning methodology which avoids forgetting knowledge gained from the source task.Compared to our prior work using EMRs from the UKY medical center, we improve both the micro and macro F-scores by more than 8%. Likewise, compared to other transfer learning methods, our approach results in nearly 2% improvement in macro F-score.RESULTSCompared to our prior work using EMRs from the UKY medical center, we improve both the micro and macro F-scores by more than 8%. Likewise, compared to other transfer learning methods, our approach results in nearly 2% improvement in macro F-score.We show that transfer learning can improve CNN performance for EMR coding in the presence of data sparsity issues. Furthermore, we find that our proposed transfer learning approach outperforms other methods with respect to macro F-score. Finally, we analyze how transfer learning impacts codes with respect to code frequency. We find that we achieve greater improvement on infrequent codes compared to improvements in most frequent codes.CONCLUSIONWe show that transfer learning can improve CNN performance for EMR coding in the presence of data sparsity issues. Furthermore, we find that our proposed transfer learning approach outperforms other methods with respect to macro F-score. Finally, we analyze how transfer learning impacts codes with respect to code frequency. We find that we achieve greater improvement on infrequent codes compared to improvements in most frequent codes.
Electronic medical records (EMRs) are manually annotated by healthcare professionals and specialized medical coders with a standardized set of alphanumeric diagnosis and procedure codes, specifically from the International Classification of Diseases (ICD). Annotating EMRs with ICD codes is important for medical billing and downstream epidemiological studies. However, manually annotating EMRs is both time-consuming and error prone. In this paper, we explore the use of convolutional neural networks (CNNs) for automatic ICD coding. Because many codes occur infrequently, CNN performance is inhibited. Therefore, we propose supplementing EMR data with PubMed indexed biomedical research abstracts through neural transfer learning. Transfer learning is the process of "transferring" knowledge acquired from one task (the source task) to a different (target) task. For the source task, we train a CNN to predict medical subject headings (MeSH) using 1.6 million PubMed indexed biomedical abstracts. For the target task, we train a CNN on 71,463 real-world EMRs collected from the University of Kentucky (UKY) medical center to predict ICD diagnosis codes. We introduce a simple, yet effective, transfer learning methodology which avoids forgetting knowledge gained from the source task. Compared to our prior work using EMRs from the UKY medical center, we improve both the micro and macro F-scores by more than 8%. Likewise, compared to other transfer learning methods, our approach results in nearly 2% improvement in macro F-score. We show that transfer learning can improve CNN performance for EMR coding in the presence of data sparsity issues. Furthermore, we find that our proposed transfer learning approach outperforms other methods with respect to macro F-score. Finally, we analyze how transfer learning impacts codes with respect to code frequency. We find that we achieve greater improvement on infrequent codes compared to improvements in most frequent codes.
Author Rios, Anthony
Kavuluru, Ramakanth
Author_xml – sequence: 1
  givenname: Anthony
  surname: Rios
  fullname: Rios, Anthony
  organization: Department of Computer Science, University of Kentucky, Lexington, KY, United States
– sequence: 2
  givenname: Ramakanth
  surname: Kavuluru
  fullname: Kavuluru, Ramakanth
  email: rvkavu2@uky.edu
  organization: Department of Computer Science, University of Kentucky, Lexington, KY, United States
BackLink https://www.ncbi.nlm.nih.gov/pubmed/31164204$$D View this record in MEDLINE/PubMed
BookMark eNqFkEtLxDAURoMoOo7-A5Eu3bTePCZtRQQRX-ADRNchTW-GjJ1Ek44w_97qqAs3swoXzvkgZ5ds-uCRkAMKBQUqj2eFjv0c24IBrQsQBQDbICNalTxnlYRNMoKa85zLSblDdlOaAUApqNwmO5xSKRiIETl7wEXUXdZH7ZPFmHWoo3d-mtkQM52Sm35frdNTH5JLmQktpqwP2eX9U9ojW1Z3Cfd_3jF5ubp8vrjJ7x6vby_O73Ij6KTPtbCsobRqoJZlxVsuQaAsUVhbc8sqzrExVNTQUGwElmChYhPZMJjoSljKx-RotfsWw_sCU6_mLhnsOu0xLJJinImaAfByQA9_0EUz5FFv0c11XKrfPw-AWAEmhpQi2j-EgvpKq2ZqlVZ9pVUg1JB20E7-acb1unfBD-1ct04-W8k4RPpwGFUyDr3B1kU0vWqDWzdw-m_AdM47o7tXXK7XPwESTqlp
CitedBy_id crossref_primary_10_1016_j_jbi_2023_104418
crossref_primary_10_3390_healthcare9111596
crossref_primary_10_1016_j_respe_2021_11_019
crossref_primary_10_1016_j_jbi_2019_103267
crossref_primary_10_1053_j_ro_2023_02_003
crossref_primary_10_1016_j_artmed_2020_101939
crossref_primary_10_1186_s12911_020_1085_4
crossref_primary_10_1016_j_imu_2023_101227
crossref_primary_10_3390_diagnostics13152582
crossref_primary_10_1186_s40537_024_00903_y
crossref_primary_10_4103_jmss_jmss_76_24
crossref_primary_10_1016_j_nic_2020_08_008
crossref_primary_10_1016_j_artmed_2023_102667
crossref_primary_10_1109_ACCESS_2023_3325896
crossref_primary_10_1109_TKDE_2022_3148267
crossref_primary_10_26599_TST_2024_9010125
crossref_primary_10_1016_j_neucom_2020_07_109
crossref_primary_10_1016_j_ijmedinf_2021_104615
crossref_primary_10_1109_ACCESS_2020_3029429
crossref_primary_10_1016_j_eswa_2022_118997
crossref_primary_10_2196_22461
crossref_primary_10_3390_app15168872
crossref_primary_10_1002_alz_12721
crossref_primary_10_1016_j_future_2021_01_013
crossref_primary_10_1007_s10489_022_04346_x
crossref_primary_10_3233_JAD_210299
crossref_primary_10_1016_j_artmed_2025_103187
Cites_doi 10.1016/j.jbi.2018.02.011
10.1561/1500000016
10.18653/v1/P18-1031
10.1007/978-3-662-44851-9_28
10.1136/amiajnl-2013-002159
10.1016/j.artmed.2015.04.007
10.1109/TPAMI.2017.2773081
10.1177/1753495X17701847
10.1016/j.neucom.2018.04.081
10.1136/jamia.2009.002733
10.1016/j.artmed.2018.03.006
10.1038/sdata.2016.35
10.1007/s10115-015-0870-3
10.1613/jair.4992
10.1093/nar/gkh061
10.1136/amiajnl-2013-002162
10.1073/pnas.1611835114
ContentType Journal Article
Copyright 2019 Elsevier B.V.
Copyright © 2019 Elsevier B.V. All rights reserved.
Copyright_xml – notice: 2019 Elsevier B.V.
– notice: Copyright © 2019 Elsevier B.V. All rights reserved.
DBID AAYXX
CITATION
CGR
CUY
CVF
ECM
EIF
NPM
7X8
DOI 10.1016/j.artmed.2019.04.002
DatabaseName CrossRef
Medline
MEDLINE
MEDLINE (Ovid)
MEDLINE
MEDLINE
PubMed
MEDLINE - Academic
DatabaseTitle CrossRef
MEDLINE
Medline Complete
MEDLINE with Full Text
PubMed
MEDLINE (Ovid)
MEDLINE - Academic
DatabaseTitleList
MEDLINE - Academic

MEDLINE
Database_xml – sequence: 1
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 2
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Medicine
Computer Science
EISSN 1873-2860
EndPage 122
ExternalDocumentID 31164204
10_1016_j_artmed_2019_04_002
S0933365718304378
Genre Research Support, Non-U.S. Gov't
Journal Article
Research Support, N.I.H., Extramural
GrantInformation_xml – fundername: NLM NIH HHS
  grantid: R21 LM012274
GroupedDBID ---
--K
--M
.1-
.DC
.FO
.~1
0R~
1B1
1P~
1RT
1~.
1~5
23N
4.4
457
4G.
53G
5GY
5VS
7-5
71M
77I
77K
8P~
9JM
9JN
AAEDT
AAEDW
AAIKJ
AAKOC
AALRI
AAOAW
AAQFI
AAQXK
AATTM
AAWTL
AAXKI
AAXUO
AAYFN
AAYWO
ABBOA
ABBQC
ABFNM
ABIVO
ABJNI
ABMAC
ABMZM
ABWVN
ABXDB
ACDAQ
ACGFS
ACIEU
ACIUM
ACLOT
ACNNM
ACRLP
ACRPL
ACVFH
ACZNC
ADBBV
ADCNI
ADEZE
ADJOM
ADMUD
ADNMO
AEBSH
AEIPS
AEKER
AENEX
AEUPX
AEVXI
AFJKZ
AFPUW
AFRHN
AFTJW
AFXIZ
AGHFR
AGQPQ
AGUBO
AGYEJ
AHHHB
AHZHX
AIALX
AIEXJ
AIGII
AIIUN
AIKHN
AITUG
AJRQY
AJUYK
AKBMS
AKRWK
AKYEP
ALMA_UNASSIGNED_HOLDINGS
AMRAJ
ANKPU
ANZVX
AOUOD
APXCP
ASPBG
AVWKF
AXJTR
AZFZN
BKOJK
BLXMC
BNPGV
CS3
EBS
EFJIC
EFKBS
EFLBG
EJD
EO8
EO9
EP2
EP3
F5P
FDB
FEDTE
FGOYB
FIRID
FNPLU
FYGXN
G-2
G-Q
GBLVA
GBOLZ
HEA
HMK
HMO
HVGLF
HZ~
IHE
J1W
KOM
LZ2
M29
M41
MO0
N9A
O-L
O9-
OAUVE
OZT
P-8
P-9
P2P
PC.
Q38
R2-
ROL
RPZ
SAE
SDF
SDG
SDP
SEL
SES
SEW
SPC
SPCBC
SSH
SSV
SSZ
T5K
UHS
WH7
WUQ
Z5R
~G-
~HD
AACTN
AAIAV
ABLVK
ABYKQ
AFCTW
AFKWA
AJBFU
AJOXV
AMFUW
LCYCR
RIG
9DU
AAYXX
CITATION
CGR
CUY
CVF
ECM
EIF
NPM
7X8
ID FETCH-LOGICAL-c415t-a4f2b118b096783d3604e67e4ff93f2833ebc1490b1eb4e70f08256b205a84f13
ISICitedReferencesCount 29
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000472704800011&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0933-3657
1873-2860
IngestDate Thu Oct 02 11:17:35 EDT 2025
Wed Feb 19 02:30:38 EST 2025
Sat Nov 29 07:04:08 EST 2025
Tue Nov 18 20:54:00 EST 2025
Fri Feb 23 02:25:16 EST 2024
Tue Oct 14 19:30:13 EDT 2025
IsPeerReviewed true
IsScholarly true
Keywords Convolutional neural networks
Transfer learning
Multi-label classification
Medical coding
Language English
License Copyright © 2019 Elsevier B.V. All rights reserved.
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c415t-a4f2b118b096783d3604e67e4ff93f2833ebc1490b1eb4e70f08256b205a84f13
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
PMID 31164204
PQID 2324920037
PQPubID 23479
PageCount 7
ParticipantIDs proquest_miscellaneous_2324920037
pubmed_primary_31164204
crossref_primary_10_1016_j_artmed_2019_04_002
crossref_citationtrail_10_1016_j_artmed_2019_04_002
elsevier_sciencedirect_doi_10_1016_j_artmed_2019_04_002
elsevier_clinicalkey_doi_10_1016_j_artmed_2019_04_002
PublicationCentury 2000
PublicationDate May 2019
2019-05-00
20190501
PublicationDateYYYYMMDD 2019-05-01
PublicationDate_xml – month: 05
  year: 2019
  text: May 2019
PublicationDecade 2010
PublicationPlace Netherlands
PublicationPlace_xml – name: Netherlands
PublicationTitle Artificial intelligence in medicine
PublicationTitleAlternate Artif Intell Med
PublicationYear 2019
Publisher Elsevier B.V
Publisher_xml – name: Elsevier B.V
References Kirkpatrick, Pascanu, Rabinowitz, Veness, Desjardins, Rusu (bib0160) 2017; 114
Aronson, Lang (bib0185) 2010; 17
Mullenbach, Wiegreffe, Duke, Sun, Eisenstein (bib0080) 2018
Glorot, Bordes, Bengio (bib0140) 2011
Johnson, Pollard, Shen, Lehman, Feng, Ghassemi (bib0010) 2016; 3
Choi, Bahadori, Schuetz, Stewart, Sun (bib0115) 2016
Kavuluru, Rios, Lu (bib0045) 2015; 65
Zeng, Li, Fei, Yu, Pan, Wang (bib0130) 2019; 324
Foo, Davis, Brown (bib0015) 2017; 10
Baumel, Nassour-Kassis, Elhadad, Elhadad (bib0035) 2017
Goldberg (bib0135) 2016; 57
Tsoumakas, Katakis, Vlahavas (bib0195) 2010
Nam, Kim, Loza Mencía, Gurevych, Fürnkranz (bib0150) 2014
Huang, Li, Pleiss, Liu, Hopcroft, Weinberger (bib0175) 2017
Liu (bib0190) 2009; 3
Sahu, Anand (bib0120) 2018; 87
Rios, Kavuluru (bib0090) 2018
Rios, Kavuluru, Lu (bib0125) 2018; 1
Bodenreider (bib0020) 2004; 32
Shi, Xie, Hu, Zhang, Xing (bib0085) 2017
Mou, Meng, Yan, Li, Xu, Zhang (bib0095) 2016
Nair, Hinton (bib0145) 2010
Al-Stouhi, Reddy (bib0100) 2016; 48
Kim (bib0030) 2014
Lee, Scott, Villarroel, Clifford, Saeed, Mark (bib0005) 2011
Li, Hoiem (bib0155) 2018; 40
Rios, Kavuluru (bib0060) 2013
Iyyer, Manjunatha, Boyd-Graber, Daumé (bib0165) 2015
Howard, Ruder (bib0105) 2018
Zeiler (bib0180) 2012
Zhang, He, Zhao, Li (bib0055) 2017; 2017
Duarte, Martins, Pinto, Silva (bib0070) 2018; 80
Oquab, Bottou, Laptev, Sivic (bib0025) 2014
Wallace, Small, Brodley, Trikalinos (bib0170) 2011
Perotte, Pivovarov, Natarajan, Weiskopf, Wood, Elhadad (bib0050) 2013; 21
Vani, Jernite, Sontag (bib0075) 2017
Wiens, Guttag, Horvitz (bib0110) 2014; 21
Pestian, Brew, Matykiewicz, Hovermale, Johnson, Cohen (bib0040) 2007
Karimi, Dai, Hassanzadeh, Nguyen (bib0065) 2017; 2017
Liu (10.1016/j.artmed.2019.04.002_bib0190) 2009; 3
Perotte (10.1016/j.artmed.2019.04.002_bib0050) 2013; 21
Baumel (10.1016/j.artmed.2019.04.002_bib0035) 2017
Kim (10.1016/j.artmed.2019.04.002_bib0030) 2014
Aronson (10.1016/j.artmed.2019.04.002_bib0185) 2010; 17
Foo (10.1016/j.artmed.2019.04.002_bib0015) 2017; 10
Huang (10.1016/j.artmed.2019.04.002_bib0175) 2017
Duarte (10.1016/j.artmed.2019.04.002_bib0070) 2018; 80
Mullenbach (10.1016/j.artmed.2019.04.002_bib0080) 2018
Shi (10.1016/j.artmed.2019.04.002_bib0085) 2017
Kirkpatrick (10.1016/j.artmed.2019.04.002_bib0160) 2017; 114
Al-Stouhi (10.1016/j.artmed.2019.04.002_bib0100) 2016; 48
Zeng (10.1016/j.artmed.2019.04.002_bib0130) 2019; 324
Nam (10.1016/j.artmed.2019.04.002_bib0150) 2014
Wiens (10.1016/j.artmed.2019.04.002_bib0110) 2014; 21
Vani (10.1016/j.artmed.2019.04.002_bib0075) 2017
Zeiler (10.1016/j.artmed.2019.04.002_bib0180) 2012
Wallace (10.1016/j.artmed.2019.04.002_bib0170) 2011
Karimi (10.1016/j.artmed.2019.04.002_bib0065) 2017; 2017
Nair (10.1016/j.artmed.2019.04.002_bib0145) 2010
Rios (10.1016/j.artmed.2019.04.002_bib0125) 2018; 1
Li (10.1016/j.artmed.2019.04.002_bib0155) 2018; 40
Oquab (10.1016/j.artmed.2019.04.002_bib0025) 2014
Kavuluru (10.1016/j.artmed.2019.04.002_bib0045) 2015; 65
Johnson (10.1016/j.artmed.2019.04.002_bib0010) 2016; 3
Howard (10.1016/j.artmed.2019.04.002_bib0105) 2018
Iyyer (10.1016/j.artmed.2019.04.002_bib0165) 2015
Goldberg (10.1016/j.artmed.2019.04.002_bib0135) 2016; 57
Lee (10.1016/j.artmed.2019.04.002_bib0005) 2011
Zhang (10.1016/j.artmed.2019.04.002_bib0055) 2017; 2017
Sahu (10.1016/j.artmed.2019.04.002_bib0120) 2018; 87
Glorot (10.1016/j.artmed.2019.04.002_bib0140) 2011
Tsoumakas (10.1016/j.artmed.2019.04.002_bib0195) 2010
Choi (10.1016/j.artmed.2019.04.002_bib0115) 2016
Pestian (10.1016/j.artmed.2019.04.002_bib0040) 2007
Rios (10.1016/j.artmed.2019.04.002_bib0090) 2018
Rios (10.1016/j.artmed.2019.04.002_bib0060) 2013
Bodenreider (10.1016/j.artmed.2019.04.002_bib0020) 2004; 32
Mou (10.1016/j.artmed.2019.04.002_bib0095) 2016
References_xml – start-page: 1746
  year: 2014
  end-page: 1751
  ident: bib0030
  article-title: Convolutional neural networks for sentence classification
  publication-title: Empirical methods in natural language processing (EMNLP)
– volume: 114
  start-page: 3521
  year: 2017
  end-page: 3526
  ident: bib0160
  article-title: Overcoming catastrophic forgetting in neural networks
  publication-title: Proc Natl Acad Sci
– volume: 3
  start-page: 225
  year: 2009
  end-page: 331
  ident: bib0190
  article-title: Learning to rank for information retrieval
  publication-title: Found Trends Inf Retriev
– volume: 80
  start-page: 64
  year: 2018
  end-page: 77
  ident: bib0070
  article-title: Deep neural models for icd-10 coding of death certificates and autopsy reports in free-text
  publication-title: J Biomed Inform
– volume: 2017
  start-page: 328
  year: 2017
  end-page: 332
  ident: bib0065
  article-title: Automatic diagnosis coding of radiology reports: a comparison of deep learning and conventional classification methods
  publication-title: BioNLP
– start-page: 437
  year: 2014
  end-page: 452
  ident: bib0150
  article-title: Large-scale multi-label text classification – revisiting neural networks
  publication-title: European conference on machine learning and knowledge discovery in databases – (ECML PKDD)
– volume: 65
  start-page: 155
  year: 2015
  end-page: 166
  ident: bib0045
  article-title: An empirical evaluation of supervised learning approaches in assigning diagnosis codes to electronic medical records
  publication-title: Artif Intell Med
– year: 2018
  ident: bib0090
  article-title: EMRcoding with semi-parametric multi-head matching networks
  publication-title: Proceedings of the North American chapter of the association for computational linguistics (NAACL)
– start-page: 97
  year: 2007
  end-page: 104
  ident: bib0040
  article-title: A shared task involving multi-label classification of clinical free text
  publication-title: Proceedings of the workshop on BioNLP: biological, translational, and clinical language processing
– start-page: 1681
  year: 2015
  end-page: 1691
  ident: bib0165
  article-title: Deep unordered composition rivals syntactic methods for text classification
  publication-title: Annual meeting of the association for computational linguistics (ACL)
– volume: 10
  start-page: 192
  year: 2017
  end-page: 194
  ident: bib0015
  article-title: Frontal lobe meningioma mimicking preeclampsia: a case study
  publication-title: Obstet Med
– volume: 57
  start-page: 345
  year: 2016
  end-page: 420
  ident: bib0135
  article-title: A primer on neural network models for natural language processing
  publication-title: J Artif Intell Res
– volume: 2017
  start-page: 263
  year: 2017
  end-page: 271
  ident: bib0055
  article-title: Enhancing automatic icd-9-cm code assignment for medical texts with pubmed
  publication-title: BioNLP
– volume: 21
  start-page: 699
  year: 2014
  end-page: 706
  ident: bib0110
  article-title: A study in transfer learning: leveraging data from multiple hospitals to enhance hospital-specific predictions
  publication-title: J Am Med Inform Assoc
– start-page: 301
  year: 2016
  end-page: 318
  ident: bib0115
  article-title: Doctor AI: predicting clinical events via recurrent neural networks
  publication-title: Machine learning for healthcare conference
– volume: 40
  start-page: 2935
  year: 2018
  end-page: 2947
  ident: bib0155
  article-title: Learning without forgetting
  publication-title: IEEE Trans Pattern Anal Mach Intell
– year: 2017
  ident: bib0175
  article-title: Snapshot ensembles: Train 1, get m for free
  publication-title: International conference on learning representations (ICLR)
– volume: 87
  start-page: 60
  year: 2018
  end-page: 66
  ident: bib0120
  article-title: What matters in a transferable neural network model for relation classification in the biomedical domain?
  publication-title: Artif Intell Med
– start-page: 754
  year: 2011
  end-page: 763
  ident: bib0170
  article-title: Class imbalance, redux
  publication-title: IEEE international conference on data mining (ICDM)
– year: 2017
  ident: bib0085
  article-title: Towards automated ICD coding using deep learning
– start-page: 328
  year: 2018
  end-page: 339
  ident: bib0105
  article-title: Universal language model fine-tuning for text classification
  publication-title: Proceedings of the 56th annual meeting of the association for computational linguistics (Volume 1: Long Papers), vol. 1
– start-page: 479
  year: 2016
  end-page: 489
  ident: bib0095
  article-title: How transferable are neural networks in NLP applications?
  publication-title: Conference on empirical methods in natural language processing (EMNLP)
– volume: 324
  start-page: 43
  year: 2019
  end-page: 50
  ident: bib0130
  article-title: Automatic icd-9 coding via deep transfer learning
  publication-title: Neurocomputing
– volume: 17
  start-page: 229
  year: 2010
  end-page: 236
  ident: bib0185
  article-title: An overview of metamap: historical perspective and recent advances
  publication-title: J Am Med Inform Assoc
– volume: 32
  start-page: D267
  year: 2004
  end-page: D270
  ident: bib0020
  article-title: The unified medical language system (umls): integrating biomedical terminology
  publication-title: Nucl Acids Res
– volume: 1
  start-page: 9
  year: 2018
  ident: bib0125
  article-title: Generalizing biomedical relation classification with neural adversarial domain adaptation
  publication-title: Bioinformatics
– volume: 3
  year: 2016
  ident: bib0010
  article-title: Mimic-iii, a freely accessible critical care database
  publication-title: Sci Data
– start-page: 667
  year: 2010
  end-page: 685
  ident: bib0195
  article-title: Mining multi-label data
  publication-title: Data mining and knowledge discovery handbook
– start-page: 8315
  year: 2011
  end-page: 8318
  ident: bib0005
  article-title: Open-access mimic-ii database for intensive care research
  publication-title: IEEE annual international conference engineering in medicine and biology society (EMBC)
– volume: 21
  start-page: 231
  year: 2013
  end-page: 237
  ident: bib0050
  article-title: Diagnosis code assignment: models and evaluation metrics
  publication-title: J Am Med Inform Assoc
– year: 2017
  ident: bib0075
  article-title: Grounded recurrent neural networks
– year: 2017
  ident: bib0035
  article-title: Multi-label classification of patient notes a case study on ICD code assignment
– start-page: 66
  year: 2013
  end-page: 73
  ident: bib0060
  article-title: Supervised extraction of diagnosis codes from EMRs: role of feature selection, data selection, and probabilistic thresholding
  publication-title: IEEE international conference on healthcare informatics (ICHI)
– volume: 48
  start-page: 201
  year: 2016
  end-page: 228
  ident: bib0100
  article-title: Transfer learning for class imbalance problems with inadequate data
  publication-title: Knowl Inf Syst
– start-page: 807
  year: 2010
  end-page: 814
  ident: bib0145
  article-title: Rectified linear units improve restricted Boltzmann machines
  publication-title: International conference on machine learning (ICML)
– year: 2012
  ident: bib0180
  article-title: ADADELTA: an adaptive learning rate method
– start-page: 1717
  year: 2014
  end-page: 1724
  ident: bib0025
  article-title: Learning and transferring mid-level image representations using convolutional neural networks
  publication-title: IEEE conference on computer vision and pattern recognition (CVPR)
– start-page: 315
  year: 2011
  end-page: 323
  ident: bib0140
  article-title: Deep sparse rectifier networks
  publication-title: International conference on artificial intelligence and statistics. JMLR W&CP Volume, vol. 15
– year: 2018
  ident: bib0080
  article-title: Explainable prediction of medical codes from clinical text
  publication-title: North American chapter of the association for computational linguistics (NAACL)
– start-page: 1717
  year: 2014
  ident: 10.1016/j.artmed.2019.04.002_bib0025
  article-title: Learning and transferring mid-level image representations using convolutional neural networks
  publication-title: IEEE conference on computer vision and pattern recognition (CVPR)
– start-page: 807
  year: 2010
  ident: 10.1016/j.artmed.2019.04.002_bib0145
  article-title: Rectified linear units improve restricted Boltzmann machines
  publication-title: International conference on machine learning (ICML)
– start-page: 1746
  year: 2014
  ident: 10.1016/j.artmed.2019.04.002_bib0030
  article-title: Convolutional neural networks for sentence classification
  publication-title: Empirical methods in natural language processing (EMNLP)
– volume: 80
  start-page: 64
  year: 2018
  ident: 10.1016/j.artmed.2019.04.002_bib0070
  article-title: Deep neural models for icd-10 coding of death certificates and autopsy reports in free-text
  publication-title: J Biomed Inform
  doi: 10.1016/j.jbi.2018.02.011
– year: 2017
  ident: 10.1016/j.artmed.2019.04.002_bib0175
  article-title: Snapshot ensembles: Train 1, get m for free
  publication-title: International conference on learning representations (ICLR)
– year: 2018
  ident: 10.1016/j.artmed.2019.04.002_bib0090
  article-title: EMRcoding with semi-parametric multi-head matching networks
  publication-title: Proceedings of the North American chapter of the association for computational linguistics (NAACL)
– start-page: 66
  year: 2013
  ident: 10.1016/j.artmed.2019.04.002_bib0060
  article-title: Supervised extraction of diagnosis codes from EMRs: role of feature selection, data selection, and probabilistic thresholding
  publication-title: IEEE international conference on healthcare informatics (ICHI)
– volume: 3
  start-page: 225
  issue: 3
  year: 2009
  ident: 10.1016/j.artmed.2019.04.002_bib0190
  article-title: Learning to rank for information retrieval
  publication-title: Found Trends Inf Retriev
  doi: 10.1561/1500000016
– start-page: 328
  year: 2018
  ident: 10.1016/j.artmed.2019.04.002_bib0105
  article-title: Universal language model fine-tuning for text classification
  publication-title: Proceedings of the 56th annual meeting of the association for computational linguistics (Volume 1: Long Papers), vol. 1
  doi: 10.18653/v1/P18-1031
– start-page: 437
  year: 2014
  ident: 10.1016/j.artmed.2019.04.002_bib0150
  article-title: Large-scale multi-label text classification – revisiting neural networks
  publication-title: European conference on machine learning and knowledge discovery in databases – (ECML PKDD)
  doi: 10.1007/978-3-662-44851-9_28
– start-page: 479
  year: 2016
  ident: 10.1016/j.artmed.2019.04.002_bib0095
  article-title: How transferable are neural networks in NLP applications?
  publication-title: Conference on empirical methods in natural language processing (EMNLP)
– start-page: 301
  year: 2016
  ident: 10.1016/j.artmed.2019.04.002_bib0115
  article-title: Doctor AI: predicting clinical events via recurrent neural networks
  publication-title: Machine learning for healthcare conference
– start-page: 8315
  year: 2011
  ident: 10.1016/j.artmed.2019.04.002_bib0005
  article-title: Open-access mimic-ii database for intensive care research
  publication-title: IEEE annual international conference engineering in medicine and biology society (EMBC)
– year: 2018
  ident: 10.1016/j.artmed.2019.04.002_bib0080
  article-title: Explainable prediction of medical codes from clinical text
  publication-title: North American chapter of the association for computational linguistics (NAACL)
– start-page: 315
  year: 2011
  ident: 10.1016/j.artmed.2019.04.002_bib0140
  article-title: Deep sparse rectifier networks
  publication-title: International conference on artificial intelligence and statistics. JMLR W&CP Volume, vol. 15
– volume: 21
  start-page: 231
  issue: 2
  year: 2013
  ident: 10.1016/j.artmed.2019.04.002_bib0050
  article-title: Diagnosis code assignment: models and evaluation metrics
  publication-title: J Am Med Inform Assoc
  doi: 10.1136/amiajnl-2013-002159
– year: 2012
  ident: 10.1016/j.artmed.2019.04.002_bib0180
– start-page: 667
  year: 2010
  ident: 10.1016/j.artmed.2019.04.002_bib0195
  article-title: Mining multi-label data
– volume: 65
  start-page: 155
  issue: 2
  year: 2015
  ident: 10.1016/j.artmed.2019.04.002_bib0045
  article-title: An empirical evaluation of supervised learning approaches in assigning diagnosis codes to electronic medical records
  publication-title: Artif Intell Med
  doi: 10.1016/j.artmed.2015.04.007
– volume: 40
  start-page: 2935
  issue: 12
  year: 2018
  ident: 10.1016/j.artmed.2019.04.002_bib0155
  article-title: Learning without forgetting
  publication-title: IEEE Trans Pattern Anal Mach Intell
  doi: 10.1109/TPAMI.2017.2773081
– volume: 2017
  start-page: 263
  year: 2017
  ident: 10.1016/j.artmed.2019.04.002_bib0055
  article-title: Enhancing automatic icd-9-cm code assignment for medical texts with pubmed
  publication-title: BioNLP
– volume: 10
  start-page: 192
  issue: 4
  year: 2017
  ident: 10.1016/j.artmed.2019.04.002_bib0015
  article-title: Frontal lobe meningioma mimicking preeclampsia: a case study
  publication-title: Obstet Med
  doi: 10.1177/1753495X17701847
– volume: 324
  start-page: 43
  year: 2019
  ident: 10.1016/j.artmed.2019.04.002_bib0130
  article-title: Automatic icd-9 coding via deep transfer learning
  publication-title: Neurocomputing
  doi: 10.1016/j.neucom.2018.04.081
– start-page: 1681
  year: 2015
  ident: 10.1016/j.artmed.2019.04.002_bib0165
  article-title: Deep unordered composition rivals syntactic methods for text classification
  publication-title: Annual meeting of the association for computational linguistics (ACL)
– volume: 17
  start-page: 229
  issue: 3
  year: 2010
  ident: 10.1016/j.artmed.2019.04.002_bib0185
  article-title: An overview of metamap: historical perspective and recent advances
  publication-title: J Am Med Inform Assoc
  doi: 10.1136/jamia.2009.002733
– volume: 87
  start-page: 60
  year: 2018
  ident: 10.1016/j.artmed.2019.04.002_bib0120
  article-title: What matters in a transferable neural network model for relation classification in the biomedical domain?
  publication-title: Artif Intell Med
  doi: 10.1016/j.artmed.2018.03.006
– start-page: 97
  year: 2007
  ident: 10.1016/j.artmed.2019.04.002_bib0040
  article-title: A shared task involving multi-label classification of clinical free text
  publication-title: Proceedings of the workshop on BioNLP: biological, translational, and clinical language processing
– start-page: 754
  year: 2011
  ident: 10.1016/j.artmed.2019.04.002_bib0170
  article-title: Class imbalance, redux
  publication-title: IEEE international conference on data mining (ICDM)
– volume: 3
  year: 2016
  ident: 10.1016/j.artmed.2019.04.002_bib0010
  article-title: Mimic-iii, a freely accessible critical care database
  publication-title: Sci Data
  doi: 10.1038/sdata.2016.35
– volume: 48
  start-page: 201
  issue: 1
  year: 2016
  ident: 10.1016/j.artmed.2019.04.002_bib0100
  article-title: Transfer learning for class imbalance problems with inadequate data
  publication-title: Knowl Inf Syst
  doi: 10.1007/s10115-015-0870-3
– volume: 57
  start-page: 345
  year: 2016
  ident: 10.1016/j.artmed.2019.04.002_bib0135
  article-title: A primer on neural network models for natural language processing
  publication-title: J Artif Intell Res
  doi: 10.1613/jair.4992
– year: 2017
  ident: 10.1016/j.artmed.2019.04.002_bib0035
– year: 2017
  ident: 10.1016/j.artmed.2019.04.002_bib0075
– volume: 2017
  start-page: 328
  year: 2017
  ident: 10.1016/j.artmed.2019.04.002_bib0065
  article-title: Automatic diagnosis coding of radiology reports: a comparison of deep learning and conventional classification methods
  publication-title: BioNLP
– volume: 32
  start-page: D267
  issue: suppl_1
  year: 2004
  ident: 10.1016/j.artmed.2019.04.002_bib0020
  article-title: The unified medical language system (umls): integrating biomedical terminology
  publication-title: Nucl Acids Res
  doi: 10.1093/nar/gkh061
– volume: 21
  start-page: 699
  issue: 4
  year: 2014
  ident: 10.1016/j.artmed.2019.04.002_bib0110
  article-title: A study in transfer learning: leveraging data from multiple hospitals to enhance hospital-specific predictions
  publication-title: J Am Med Inform Assoc
  doi: 10.1136/amiajnl-2013-002162
– volume: 1
  start-page: 9
  year: 2018
  ident: 10.1016/j.artmed.2019.04.002_bib0125
  article-title: Generalizing biomedical relation classification with neural adversarial domain adaptation
  publication-title: Bioinformatics
– year: 2017
  ident: 10.1016/j.artmed.2019.04.002_bib0085
– volume: 114
  start-page: 3521
  issue: 13
  year: 2017
  ident: 10.1016/j.artmed.2019.04.002_bib0160
  article-title: Overcoming catastrophic forgetting in neural networks
  publication-title: Proc Natl Acad Sci
  doi: 10.1073/pnas.1611835114
SSID ssj0007416
Score 2.3971834
Snippet •Transfer learning using convolutional neural networks improves multi-label learning.•Predicting MeSH terms for biomedical articles is a useful source task for...
Electronic medical records (EMRs) are manually annotated by healthcare professionals and specialized medical coders with a standardized set of alphanumeric...
SourceID proquest
pubmed
crossref
elsevier
SourceType Aggregation Database
Index Database
Enrichment Source
Publisher
StartPage 116
SubjectTerms Clinical Coding - methods
Convolutional neural networks
Electronic Health Records - organization & administration
Electronic Health Records - standards
Humans
International Classification of Diseases
Medical coding
Multi-label classification
Neural Networks, Computer
Transfer learning
Title Neural transfer learning for assigning diagnosis codes to EMRs
URI https://www.clinicalkey.com/#!/content/1-s2.0-S0933365718304378
https://dx.doi.org/10.1016/j.artmed.2019.04.002
https://www.ncbi.nlm.nih.gov/pubmed/31164204
https://www.proquest.com/docview/2324920037
Volume 96
WOSCitedRecordID wos000472704800011&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVESC
  databaseName: Elsevier SD Freedom Collection Journals 2021
  customDbUrl:
  eissn: 1873-2860
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0007416
  issn: 0933-3657
  databaseCode: AIEXJ
  dateStart: 19950201
  isFulltext: true
  titleUrlDefault: https://www.sciencedirect.com
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1ba9swFBZrO8Zedulu2aVoMPYyXGxLjuWXQRgp22izUVLIm5BsiaXrnMyOS3_-jm5uICvNHvYiHGEljr7PR-dI54LQu4IyyipBIl2SLKI0rSImhIaGCJWkksiqssUm8smEzWbFd19WsbXlBPK6ZldXxfK_Qg19ALYJnf0HuPsvhQ64BtChBdih3Qp4k27DepAbjVQ1oS6E95dsjcOG-VQ5H7t5-8FEtds8D-OT03ZdWR011pHIlvVYz9w5rzeO5E_nzl3P5yLoxbi47C66prNAil_iJ-D4Y32jwcQ2Bbc-LxtZTqKUufT_h-ovfV6gFusSMXGhlBuS2m0anB_CPMEjGx-7wuacjdPrlSmcxk--8aOz42M-Hc-m75e_I1MzzJyt-wIqO2gvzbMCxPLe6Mt49rVfiY22aQzu8JAhdNL6923-8E2qyU2mh1VBpo_QA2874JHD_DG6o-p99DDU5cBeTO-jeycenSfoo2MDDmzAgQ0Y2IB7NuCeDdiyAa8W2LDhKTo7Gk8_fY58yYyoBE1sFQmqUwk2owTLNGekIsOYqmGuqNYF0aBKEiVLMIpjmShJVR5rs0UwlGmcCUZ1Qp6h3XpRqxcIxyWTIstkoRNFK7iQWlVMJIwVmQTLfYBImC1e-nzypqzJBQ-Og-fczTE3c8xjymGOByjqRy1dPpVb7s8CEDzECsPqxoFGt4zL-3Fel3Q64hYj3wa8OYhac34marXoWm6Mj8I4c-YD9NwRof8PBHhO05i-3GL0K3T_-v16jXZXTafeoLvl5WreNgdoJ5-xA0_lP_QXqfw
linkProvider Elsevier
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Neural+transfer+learning+for+assigning+diagnosis+codes+to+EMRs&rft.jtitle=Artificial+intelligence+in+medicine&rft.au=Rios%2C+Anthony&rft.au=Kavuluru%2C+Ramakanth&rft.date=2019-05-01&rft.issn=1873-2860&rft.eissn=1873-2860&rft.volume=96&rft.spage=116&rft_id=info:doi/10.1016%2Fj.artmed.2019.04.002&rft.externalDBID=NO_FULL_TEXT
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0933-3657&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0933-3657&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0933-3657&client=summon