MEDRank: Using graph-based concept ranking to index biomedical texts
► We define, implement and evaluate MEDRank. ► MEDRank is a graph-based algorithm that identifies important concepts in text. ► MEDRank improves retrieval of major Medical Subject Headings by 30%. ► Terms selected by MEDRank are match human expectations better than alternatives. As the volume of bio...
Saved in:
| Published in: | International journal of medical informatics (Shannon, Ireland) Vol. 80; no. 6; pp. 431 - 441 |
|---|---|
| Main Authors: | , , , , , |
| Format: | Journal Article |
| Language: | English |
| Published: |
Ireland
Elsevier Ireland Ltd
01.06.2011
|
| Subjects: | |
| ISSN: | 1386-5056, 1872-8243, 1872-8243 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract | ► We define, implement and evaluate MEDRank. ► MEDRank is a graph-based algorithm that identifies important concepts in text. ► MEDRank improves retrieval of major Medical Subject Headings by 30%. ► Terms selected by MEDRank are match human expectations better than alternatives.
As the volume of biomedical text increases exponentially, automatic indexing becomes increasingly important. However, existing approaches do not distinguish central (or core) concepts from concepts that were mentioned in passing. We focus on the problem of indexing MEDLINE records, a process that is currently performed by highly trained humans at the National Library of Medicine (NLM). NLM indexers are assisted by a system called the Medical Text Indexer (MTI) that suggests candidate indexing terms.
To improve the ability of MTI to select the core terms in MEDLINE abstracts. These core concepts are deemed to be most important and are designated as “major headings” by MEDLINE indexers. We introduce and evaluate a graph-based indexing methodology called MEDRank that generates concept graphs from biomedical text and then ranks the concepts within these graphs to identify the most important ones.
We insert a MEDRank step into the MTI and compare MTI's output with and without MEDRank to the MEDLINE indexers’ selected terms for a sample of 11,803 PubMed Central articles. We also tested whether human raters prefer terms generated by the MEDLINE indexers, MTI without MEDRank, and MTI with MEDRank for a sample of 36 PubMed Central articles.
MEDRank improved recall of major headings designated by 30% over MTI without MEDRank (0.489 vs. 0.376). Overall recall was only slightly (6.5%) higher (0.490 vs. 0.460) as was F
2 (3%, 0.408 vs. 0.396). However, overall precision was 3.9% lower (0.268 vs. 0.279). Human raters preferred terms generated by MTI with MEDRank over terms generated by MTI without MEDRank (by an average of 1.00 more term per article), and preferred terms generated by MTI with MEDRank and the MEDLINE indexers at the same rate.
The addition of MEDRank to MTI significantly improved the retrieval of core concepts in MEDLINE abstracts and more closely matched human expectations compared to MTI without MEDRank. In addition, MEDRank slightly improved overall recall and F
2. |
|---|---|
| AbstractList | ► We define, implement and evaluate MEDRank. ► MEDRank is a graph-based algorithm that identifies important concepts in text. ► MEDRank improves retrieval of major Medical Subject Headings by 30%. ► Terms selected by MEDRank are match human expectations better than alternatives.
As the volume of biomedical text increases exponentially, automatic indexing becomes increasingly important. However, existing approaches do not distinguish central (or core) concepts from concepts that were mentioned in passing. We focus on the problem of indexing MEDLINE records, a process that is currently performed by highly trained humans at the National Library of Medicine (NLM). NLM indexers are assisted by a system called the Medical Text Indexer (MTI) that suggests candidate indexing terms.
To improve the ability of MTI to select the core terms in MEDLINE abstracts. These core concepts are deemed to be most important and are designated as “major headings” by MEDLINE indexers. We introduce and evaluate a graph-based indexing methodology called MEDRank that generates concept graphs from biomedical text and then ranks the concepts within these graphs to identify the most important ones.
We insert a MEDRank step into the MTI and compare MTI's output with and without MEDRank to the MEDLINE indexers’ selected terms for a sample of 11,803 PubMed Central articles. We also tested whether human raters prefer terms generated by the MEDLINE indexers, MTI without MEDRank, and MTI with MEDRank for a sample of 36 PubMed Central articles.
MEDRank improved recall of major headings designated by 30% over MTI without MEDRank (0.489 vs. 0.376). Overall recall was only slightly (6.5%) higher (0.490 vs. 0.460) as was F
2 (3%, 0.408 vs. 0.396). However, overall precision was 3.9% lower (0.268 vs. 0.279). Human raters preferred terms generated by MTI with MEDRank over terms generated by MTI without MEDRank (by an average of 1.00 more term per article), and preferred terms generated by MTI with MEDRank and the MEDLINE indexers at the same rate.
The addition of MEDRank to MTI significantly improved the retrieval of core concepts in MEDLINE abstracts and more closely matched human expectations compared to MTI without MEDRank. In addition, MEDRank slightly improved overall recall and F
2. Highlights ► We define, implement and evaluate MEDRank. ► MEDRank is a graph-based algorithm that identifies important concepts in text. ► MEDRank improves retrieval of major Medical Subject Headings by 30%. ► Terms selected by MEDRank are match human expectations better than alternatives. As the volume of biomedical text increases exponentially, automatic indexing becomes increasingly important. However, existing approaches do not distinguish central (or core) concepts from concepts that were mentioned in passing. We focus on the problem of indexing MEDLINE records, a process that is currently performed by highly trained humans at the National Library of Medicine (NLM). NLM indexers are assisted by a system called the Medical Text Indexer (MTI) that suggests candidate indexing terms. To improve the ability of MTI to select the core terms in MEDLINE abstracts. These core concepts are deemed to be most important and are designated as "major headings" by MEDLINE indexers. We introduce and evaluate a graph-based indexing methodology called MEDRank that generates concept graphs from biomedical text and then ranks the concepts within these graphs to identify the most important ones. We insert a MEDRank step into the MTI and compare MTI's output with and without MEDRank to the MEDLINE indexers' selected terms for a sample of 11,803 PubMed Central articles. We also tested whether human raters prefer terms generated by the MEDLINE indexers, MTI without MEDRank, and MTI with MEDRank for a sample of 36 PubMed Central articles. MEDRank improved recall of major headings designated by 30% over MTI without MEDRank (0.489 vs. 0.376). Overall recall was only slightly (6.5%) higher (0.490 vs. 0.460) as was F(2) (3%, 0.408 vs. 0.396). However, overall precision was 3.9% lower (0.268 vs. 0.279). Human raters preferred terms generated by MTI with MEDRank over terms generated by MTI without MEDRank (by an average of 1.00 more term per article), and preferred terms generated by MTI with MEDRank and the MEDLINE indexers at the same rate. The addition of MEDRank to MTI significantly improved the retrieval of core concepts in MEDLINE abstracts and more closely matched human expectations compared to MTI without MEDRank. In addition, MEDRank slightly improved overall recall and F(2). As the volume of biomedical text increases exponentially, automatic indexing becomes increasingly important. However, existing approaches do not distinguish central (or core) concepts from concepts that were mentioned in passing. We focus on the problem of indexing MEDLINE records, a process that is currently performed by highly trained humans at the National Library of Medicine (NLM). NLM indexers are assisted by a system called the Medical Text Indexer (MTI) that suggests candidate indexing terms.BACKGROUNDAs the volume of biomedical text increases exponentially, automatic indexing becomes increasingly important. However, existing approaches do not distinguish central (or core) concepts from concepts that were mentioned in passing. We focus on the problem of indexing MEDLINE records, a process that is currently performed by highly trained humans at the National Library of Medicine (NLM). NLM indexers are assisted by a system called the Medical Text Indexer (MTI) that suggests candidate indexing terms.To improve the ability of MTI to select the core terms in MEDLINE abstracts. These core concepts are deemed to be most important and are designated as "major headings" by MEDLINE indexers. We introduce and evaluate a graph-based indexing methodology called MEDRank that generates concept graphs from biomedical text and then ranks the concepts within these graphs to identify the most important ones.OBJECTIVETo improve the ability of MTI to select the core terms in MEDLINE abstracts. These core concepts are deemed to be most important and are designated as "major headings" by MEDLINE indexers. We introduce and evaluate a graph-based indexing methodology called MEDRank that generates concept graphs from biomedical text and then ranks the concepts within these graphs to identify the most important ones.We insert a MEDRank step into the MTI and compare MTI's output with and without MEDRank to the MEDLINE indexers' selected terms for a sample of 11,803 PubMed Central articles. We also tested whether human raters prefer terms generated by the MEDLINE indexers, MTI without MEDRank, and MTI with MEDRank for a sample of 36 PubMed Central articles.METHODSWe insert a MEDRank step into the MTI and compare MTI's output with and without MEDRank to the MEDLINE indexers' selected terms for a sample of 11,803 PubMed Central articles. We also tested whether human raters prefer terms generated by the MEDLINE indexers, MTI without MEDRank, and MTI with MEDRank for a sample of 36 PubMed Central articles.MEDRank improved recall of major headings designated by 30% over MTI without MEDRank (0.489 vs. 0.376). Overall recall was only slightly (6.5%) higher (0.490 vs. 0.460) as was F(2) (3%, 0.408 vs. 0.396). However, overall precision was 3.9% lower (0.268 vs. 0.279). Human raters preferred terms generated by MTI with MEDRank over terms generated by MTI without MEDRank (by an average of 1.00 more term per article), and preferred terms generated by MTI with MEDRank and the MEDLINE indexers at the same rate.RESULTSMEDRank improved recall of major headings designated by 30% over MTI without MEDRank (0.489 vs. 0.376). Overall recall was only slightly (6.5%) higher (0.490 vs. 0.460) as was F(2) (3%, 0.408 vs. 0.396). However, overall precision was 3.9% lower (0.268 vs. 0.279). Human raters preferred terms generated by MTI with MEDRank over terms generated by MTI without MEDRank (by an average of 1.00 more term per article), and preferred terms generated by MTI with MEDRank and the MEDLINE indexers at the same rate.The addition of MEDRank to MTI significantly improved the retrieval of core concepts in MEDLINE abstracts and more closely matched human expectations compared to MTI without MEDRank. In addition, MEDRank slightly improved overall recall and F(2).CONCLUSIONSThe addition of MEDRank to MTI significantly improved the retrieval of core concepts in MEDLINE abstracts and more closely matched human expectations compared to MTI without MEDRank. In addition, MEDRank slightly improved overall recall and F(2). As the volume of biomedical text increases exponentially, automatic indexing becomes increasingly important. However, existing approaches do not distinguish central (or core) concepts from concepts that were mentioned in passing. We focus on the problem of indexing MEDLINE records, a process that is currently performed by highly trained humans at the National Library of Medicine (NLM). NLM indexers are assisted by a system called the Medical Text Indexer (MTI) that suggests candidate indexing terms. Objective: To improve the ability of MTI to select the core terms in MEDLINE abstracts. These core concepts are deemed to be most important and are designated as "major headings" by MEDLINE indexers. We introduce and evaluate a graph-based indexing methodology called MEDRank that generates concept graphs from biomedical text and then ranks the concepts within these graphs to identify the most important ones. Methods: We insert a MEDRank step into the MTI and compare MTI's output with and without MEDRank to the MEDLINE indexers' selected terms for a sample of 11,803 PubMed Central articles. We also tested whether human raters prefer terms generated by the MEDLINE indexers, MTI without MEDRank, and MTI with MEDRank for a sample of 36 PubMed Central articles. Results: MEDRank improved recall of major headings designated by 30% over MTI without MEDRank (0.489 vs. 0.376). Overall recall was only slightly (6.5%) higher (0.490 vs. 0.460) as was F sub(2 (3%, 0.408 vs. 0.396). However, overall precision was 3.9% lower (0.268 vs. 0.279). Human raters preferred terms generated by MTI with MEDRank over terms generated by MTI without MEDRank (by an average of 1.00 more term per article), and preferred terms generated by MTI with MEDRank and the MEDLINE indexers at the same rate. Conclusions: The addition of MEDRank to MTI significantly improved the retrieval of core concepts in MEDLINE abstracts and more closely matched human expectations compared to MTI without MEDRank. In addition, MEDRank slightly improved overall recall and F) sub(2). |
| Author | Cohen, Trevor Herskovic, Jorge R. Bernstam, Elmer V. Iyengar, M. Sriram Smith, Jack W. Subramanian, Devika |
| AuthorAffiliation | 2 Rice University Engineering School, Department of Computer Science 4 Department of Internal Medicine, Medical School, The University of Texas Health Science Center at Houston 1 School of Biomedical Informatics, The University of Texas Health Science Center at Houston 3 NASA Johnson Space Center |
| AuthorAffiliation_xml | – name: 4 Department of Internal Medicine, Medical School, The University of Texas Health Science Center at Houston – name: 1 School of Biomedical Informatics, The University of Texas Health Science Center at Houston – name: 2 Rice University Engineering School, Department of Computer Science – name: 3 NASA Johnson Space Center |
| Author_xml | – sequence: 1 givenname: Jorge R. surname: Herskovic fullname: Herskovic, Jorge R. organization: School of Biomedical Informatics, The University of Texas Health Science Center at Houston, United States – sequence: 2 givenname: Trevor surname: Cohen fullname: Cohen, Trevor organization: School of Biomedical Informatics, The University of Texas Health Science Center at Houston, United States – sequence: 3 givenname: Devika surname: Subramanian fullname: Subramanian, Devika organization: Rice University Engineering School, Department of Computer Science, United States – sequence: 4 givenname: M. Sriram surname: Iyengar fullname: Iyengar, M. Sriram organization: School of Biomedical Informatics, The University of Texas Health Science Center at Houston, United States – sequence: 5 givenname: Jack W. surname: Smith fullname: Smith, Jack W. organization: School of Biomedical Informatics, The University of Texas Health Science Center at Houston, United States – sequence: 6 givenname: Elmer V. surname: Bernstam fullname: Bernstam, Elmer V. email: elmer.v.bernstam@uth.tmc.edu organization: School of Biomedical Informatics, The University of Texas Health Science Center at Houston, United States |
| BackLink | https://www.ncbi.nlm.nih.gov/pubmed/21439897$$D View this record in MEDLINE/PubMed |
| BookMark | eNqNkltv1DAQhSNURC_wF6q88ZQw49wchCpQW1qkIiSgz5bjTLZOs_bW9lbtv8fRdhH0geXJluac4yN_c5jsGWsoSY4RcgSs3425HpfUazPkDBBzYDkAf5EcIG9YxllZ7MV7weusgqreTw69HwGwgap8lewzLIuWt81Bcvb1_Oy7NLfv02uvzSJdOLm6yTrpqU-VNYpWIXVxPs-CTbXp6SHttJ3fVnJKAz0E_zp5OcjJ05un8yi5_nz-8_Qyu_p28eX001WmqroMsRUhEecdYcmgr_qSY1cWXcGBmkqpjtfUFS1K7LqGKsYqRGwHbHritRxYcZScbHJX6y4WUGSCk5NYOb2U7lFYqcXfE6NvxMLeiwJaqHkbA94-BTh7tyYfxFJ7RdMkDdm1F7wpsQXOYLeyLhsOWNRRefxnqd9ttn8cBR82AuWs944GoXSQQdu5o54EgpiRilFskYoZqQAmItJor5_Zty_sNH7cGCkiudfkhFeaItJeO1JB9Fbvjjh5FqEmbWbwt_RIfrRrZyJwgcJHg_gxL9y8b4gAcdPYvwP-p8EvKKrpKQ |
| CitedBy_id | crossref_primary_10_1186_s12859_015_0539_7 crossref_primary_10_1016_j_jbi_2012_11_001 crossref_primary_10_1186_1471_2105_13_S13_S2 crossref_primary_10_1186_s12911_020_01330_8 crossref_primary_10_1007_s11042_023_14613_9 crossref_primary_10_1016_j_jksuci_2024_102178 crossref_primary_10_1186_1471_2105_14_113 crossref_primary_10_1371_journal_pone_0209961 crossref_primary_10_1016_j_ijmedinf_2020_104161 |
| Cites_doi | 10.1197/jamia.M1909 10.1086/392651 10.1016/j.jbi.2006.06.004 10.1016/j.jbi.2008.12.007 10.1016/j.jbi.2009.02.002 10.1016/j.jbi.2009.09.003 |
| ContentType | Journal Article |
| Copyright | 2011 Elsevier Ireland Ltd Elsevier Ireland Ltd Copyright © 2011 Elsevier Ireland Ltd. All rights reserved. 2011 Elsevier Ireland Ltd. All rights reserved. 2011 |
| Copyright_xml | – notice: 2011 Elsevier Ireland Ltd – notice: Elsevier Ireland Ltd – notice: Copyright © 2011 Elsevier Ireland Ltd. All rights reserved. – notice: 2011 Elsevier Ireland Ltd. All rights reserved. 2011 |
| DBID | AAYXX CITATION CGR CUY CVF ECM EIF NPM 7X8 7QO 8FD FR3 P64 5PM |
| DOI | 10.1016/j.ijmedinf.2011.02.008 |
| DatabaseName | CrossRef Medline MEDLINE MEDLINE (Ovid) MEDLINE MEDLINE PubMed MEDLINE - Academic Biotechnology Research Abstracts Technology Research Database Engineering Research Database Biotechnology and BioEngineering Abstracts PubMed Central (Full Participant titles) |
| DatabaseTitle | CrossRef MEDLINE Medline Complete MEDLINE with Full Text PubMed MEDLINE (Ovid) MEDLINE - Academic Engineering Research Database Biotechnology Research Abstracts Technology Research Database Biotechnology and BioEngineering Abstracts |
| DatabaseTitleList | MEDLINE MEDLINE - Academic Engineering Research Database |
| 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 |
| EISSN | 1872-8243 |
| EndPage | 441 |
| ExternalDocumentID | PMC3090689 21439897 10_1016_j_ijmedinf_2011_02_008 S1386505611000542 1_s2_0_S1386505611000542 |
| Genre | Journal Article Research Support, N.I.H., Extramural |
| GeographicLocations | United States |
| GeographicLocations_xml | – name: United States |
| GrantInformation_xml | – fundername: NCRR NIH HHS grantid: RC1 RR028254 – fundername: CCR NIH HHS grantid: 1RC1RR028254 – fundername: NLM NIH HHS grantid: K22 LM008306 – fundername: NCRR NIH HHS grantid: 3UL1RR024148 – fundername: NLM NIH HHS grantid: 5K22LM8306 – fundername: NCRR NIH HHS grantid: UL1 RR024148 – fundername: National Center for Research Resources : NCRR grantid: RC1 RR028254-02 || RR – fundername: National Center for Research Resources : NCRR grantid: UL1 RR024148-01 || RR – fundername: National Center for Research Resources : NCRR grantid: RC1 RR028254-01 || RR – fundername: National Library of Medicine : NLM grantid: K22 LM008306-01 || LM – fundername: National Library of Medicine : NLM grantid: K22 LM008306-02 || LM – fundername: National Library of Medicine : NLM grantid: K22 LM008306-03 || LM |
| GroupedDBID | --- --K --M -~X .1- .FO .GJ .~1 0R~ 1B1 1P~ 1RT 1~. 1~5 29J 4.4 457 4G. 53G 5GY 5VS 7-5 71M 8P~ AABNK AAEDT AAEDW AAIKJ AAKOC AALRI AAOAW AAQFI AAQXK AATTM AAWTL AAXKI AAXUO AAYFN AAYWO ABBOA ABBQC ABDPE ABFNM ABJNI ABMAC ABMZM ABWVN ABXDB ACDAQ ACGFS ACIEU ACIUM ACJTP ACLOT ACNNM ACRLP ACRPL ACVFH ACZNC ADBBV ADCNI ADEZE ADJOM ADMUD ADNMO AEBSH AEIPS AEKER AENEX AEUPX AEVXI AFJKZ AFPUW AFRHN AFTJW AFXBA AFXIZ AGHFR AGQPQ AGUBO AGYEJ 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 DU5 EBS EFJIC EFKBS EFLBG EJD EO8 EO9 EP2 EP3 F5P FDB FEDTE FGOYB FIRID FNPLU FYGXN G-Q GBLVA GBOLZ HVGLF HZ~ IHE J1W KOM M41 MO0 N9A O-L O9- OAUVE OZT P-8 P-9 P2P PC. Q38 R2- ROL RPZ SCC SDF SDG SDP SEL SES SEW SNG SPC SPCBC SSH SSV SSZ T5K UHS Z5R ~G- ~HD AACTN AFCTW AFKWA AJOXV AMFUW RIG AAIAV ABLVK ABYKQ AISVY AJBFU G8K LCYCR NAHTW 9DU AAYXX CITATION AGCQF AGRNS CGR CUY CVF ECM EIF NPM 7X8 7QO 8FD FR3 P64 5PM |
| ID | FETCH-LOGICAL-c564t-82e1ee88be1420d5d481b43b380e75ccb86eb391a1bb7e52251119f17de86af23 |
| ISICitedReferencesCount | 10 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000290017100006&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1386-5056 1872-8243 |
| IngestDate | Tue Sep 30 17:16:15 EDT 2025 Tue Oct 07 09:33:01 EDT 2025 Sun Nov 09 09:53:11 EST 2025 Mon Jul 21 06:04:02 EDT 2025 Sat Nov 29 04:07:09 EST 2025 Tue Nov 18 22:21:29 EST 2025 Fri Feb 23 02:22:43 EST 2024 Sun Feb 23 10:19:40 EST 2025 Tue Oct 14 19:29:47 EDT 2025 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 6 |
| Keywords | MEDLINE Automatic data processing Algorithms Abstracting and indexing as topic Natural language processing PubMed Digital libraries Medical informatics |
| Language | English |
| License | https://www.elsevier.com/tdm/userlicense/1.0 Copyright © 2011 Elsevier Ireland Ltd. All rights reserved. |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-c564t-82e1ee88be1420d5d481b43b380e75ccb86eb391a1bb7e52251119f17de86af23 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 ObjectType-Article-2 ObjectType-Feature-1 |
| PMID | 21439897 |
| PQID | 864780136 |
| PQPubID | 23479 |
| PageCount | 11 |
| ParticipantIDs | pubmedcentral_primary_oai_pubmedcentral_nih_gov_3090689 proquest_miscellaneous_874190820 proquest_miscellaneous_864780136 pubmed_primary_21439897 crossref_citationtrail_10_1016_j_ijmedinf_2011_02_008 crossref_primary_10_1016_j_ijmedinf_2011_02_008 elsevier_sciencedirect_doi_10_1016_j_ijmedinf_2011_02_008 elsevier_clinicalkeyesjournals_1_s2_0_S1386505611000542 elsevier_clinicalkey_doi_10_1016_j_ijmedinf_2011_02_008 |
| PublicationCentury | 2000 |
| PublicationDate | 2011-06-01 |
| PublicationDateYYYYMMDD | 2011-06-01 |
| PublicationDate_xml | – month: 06 year: 2011 text: 2011-06-01 day: 01 |
| PublicationDecade | 2010 |
| PublicationPlace | Ireland |
| PublicationPlace_xml | – name: Ireland |
| PublicationTitle | International journal of medical informatics (Shannon, Ireland) |
| PublicationTitleAlternate | Int J Med Inform |
| PublicationYear | 2011 |
| Publisher | Elsevier Ireland Ltd |
| Publisher_xml | – name: Elsevier Ireland Ltd |
| References | Cohen, Widdows (bib0115) 2009; 42 Vasuki, Cohen (bib0040) 2010; 43 Cohen, Schvaneveldt, Widdows (bib0130) 2010; 43 Pedersen, Pakhomov, Patwardhan, Chute (bib0120) 2007; 40 Ruiz ME, Aronson AR. User-centered Evaluation of the Medical Text Indexing (MTI) system. U.S. National Library of Medicine (2007). (cited 21.12.09) (2009). U.S. National Library of Medicine. MeSH history. (Web page) Bethesda, MD: National Library of Medicine. Available from U.S. National Library of Medicine. PubMed Related Citations algorithm. Available from U.S. National Library of Medicine. Restrict to MeSH algorithm March 16, 2008 (November 6) (2004). U.S. National Library of Medicine. Medical Subject Heading (MESH) Fact Sheet. (Web page) Bethesda, MD: U.S. National Library of Medicine. Available from Suppe (bib0045) 1998; 65 Gay, Kayaalp, Aronson (bib0025) 2005 Bernstam, Herskovic, Aphinyanaphongs, Aliferis, Sriram, Hersh (bib0065) 2006; 13 U.S. National Library of Medicine. Trigram Algorithm. Available from Page L, Brin S, Motwani R, Winograd T. The PageRank Citation Ranking: Bringing Order to the Web. Stanford Publications. Available from U.S. National Library of Medicine. MetaMap Indexing Algorithm. Available from U.S. National Library of Medicine. Online services reference manual. Bethesda, MD (1982). Aronson, Mork, Gay, Humphrey, Rogers (bib0145) 2004; 107 Névéol, Shooshan, Humphrey, Mork, Aronson (bib0005) 2008; 42 Fiszman, Rindflesch, Kilicoglu (bib0060) 2004 Bondy, Murty (bib0055) 1976 (updated February 23, cited 18.04.07) (2007). Pujol, Sanguesa, Delgado (bib0070) 2002 U.S. National Library of Medicine. Clustering and Ranking process. Available from (bib0125) 2000 (bib0165) 1998 Kim, Aronson, Wilbur (bib0030) 2001 (bib0075) 2004 Kintsch (bib0050) 1998 (updated March 16, cited 29.12.09) (2004). (updated October 30, 2007, cited 25.09.08) (1999). U.S. National Library of Medicine. Medical Text Indexer (MTI). Bethesda, MD. Available from (updated March 16, 06.11.08) (2004). Gay (bib0160) 2006 (updated March 16, cited 06.11.08) (2004). Funk, Reid, McGoogan (bib0020) 1983; 71 U.S. National Library of Medicine. Principles of MEDLINE Subject Indexing. (Web page) Bethesda, MD: National Library of Medicine. Available from Hecht-Nielsen (bib0135) 1994 (updated March 16, cited 6.11.08) (2004). Lehmann, Lehmann, Erich (bib0155) 2006 Kim, Aronson, Mork, Cohen, Lehmann (bib0175) 2004; 107 (updated November 27, cited 30.03.07) (2006). (bib0140) 2005 (bib0185) 2009 (bib0170) 2007 (15.03.05) (1998). Kim (10.1016/j.ijmedinf.2011.02.008_bib0175) 2004; 107 10.1016/j.ijmedinf.2011.02.008_bib0100 Aronson (10.1016/j.ijmedinf.2011.02.008_bib0145) 2004; 107 Gay (10.1016/j.ijmedinf.2011.02.008_bib0025) 2005 (10.1016/j.ijmedinf.2011.02.008_bib0125) 2000 10.1016/j.ijmedinf.2011.02.008_bib0105 Cohen (10.1016/j.ijmedinf.2011.02.008_bib0115) 2009; 42 Kintsch (10.1016/j.ijmedinf.2011.02.008_bib0050) 1998 Funk (10.1016/j.ijmedinf.2011.02.008_bib0020) 1983; 71 Suppe (10.1016/j.ijmedinf.2011.02.008_bib0045) 1998; 65 (10.1016/j.ijmedinf.2011.02.008_bib0185) 2009 Hecht-Nielsen (10.1016/j.ijmedinf.2011.02.008_bib0135) 1994 Névéol (10.1016/j.ijmedinf.2011.02.008_bib0005) 2008; 42 Vasuki (10.1016/j.ijmedinf.2011.02.008_bib0040) 2010; 43 10.1016/j.ijmedinf.2011.02.008_bib0095 10.1016/j.ijmedinf.2011.02.008_bib0150 (10.1016/j.ijmedinf.2011.02.008_bib0165) 1998 10.1016/j.ijmedinf.2011.02.008_bib0010 Pedersen (10.1016/j.ijmedinf.2011.02.008_bib0120) 2007; 40 Cohen (10.1016/j.ijmedinf.2011.02.008_bib0130) 2010; 43 10.1016/j.ijmedinf.2011.02.008_bib0090 Kim (10.1016/j.ijmedinf.2011.02.008_bib0030) 2001 10.1016/j.ijmedinf.2011.02.008_bib0110 10.1016/j.ijmedinf.2011.02.008_bib0035 (10.1016/j.ijmedinf.2011.02.008_bib0170) 2007 10.1016/j.ijmedinf.2011.02.008_bib0015 (10.1016/j.ijmedinf.2011.02.008_bib0140) 2005 Gay (10.1016/j.ijmedinf.2011.02.008_bib0160) 2006 (10.1016/j.ijmedinf.2011.02.008_bib0075) 2004 10.1016/j.ijmedinf.2011.02.008_bib0080 10.1016/j.ijmedinf.2011.02.008_bib0180 Bondy (10.1016/j.ijmedinf.2011.02.008_bib0055) 1976 Lehmann (10.1016/j.ijmedinf.2011.02.008_bib0155) 2006 10.1016/j.ijmedinf.2011.02.008_bib0085 Pujol (10.1016/j.ijmedinf.2011.02.008_bib0070) 2002 Fiszman (10.1016/j.ijmedinf.2011.02.008_bib0060) 2004 Bernstam (10.1016/j.ijmedinf.2011.02.008_bib0065) 2006; 13 |
| References_xml | – reference: U.S. National Library of Medicine. Restrict to MeSH algorithm March 16, 2008 (November 6) (2004). – reference: U.S. National Library of Medicine. Principles of MEDLINE Subject Indexing. (Web page) Bethesda, MD: National Library of Medicine. Available from: – reference: Page L, Brin S, Motwani R, Winograd T. The PageRank Citation Ranking: Bringing Order to the Web. Stanford Publications. Available from: – reference: U.S. National Library of Medicine. Clustering and Ranking process. Available from: – reference: Ruiz ME, Aronson AR. User-centered Evaluation of the Medical Text Indexing (MTI) system. U.S. National Library of Medicine (2007). – year: 1976 ident: bib0055 article-title: Graph Theory with Applications – volume: 13 start-page: 96 year: 2006 end-page: 105 ident: bib0065 article-title: Using citation data to improve retrieval from MEDLINE publication-title: J. Am. Med. Inform. Assoc. – reference: , (updated October 30, 2007, cited 25.09.08) (1999). – reference: , (updated March 16, 06.11.08) (2004). – year: 2006 ident: bib0155 article-title: Nonparametrics publication-title: Statistical Methods Based on Ranks. With the Special Assistance – start-page: 319 year: 2001 end-page: 323 ident: bib0030 article-title: Automatic MeSH term assignment and quality assessment publication-title: Proc. AMIA Symp. – year: 1994 ident: bib0135 article-title: Context vectors; general purpose approximate meaning representations self-organized from raw data publication-title: Computational Intelligence: Imitating Life – volume: 107 start-page: 268 year: 2004 end-page: 272 ident: bib0145 article-title: The NLM Indexing Initiative's Medical Text Indexer publication-title: Stud. Health Technol. Inform. – year: 1998 ident: bib0165 article-title: KeyGraph: automatic indexing by co-occurrence graph based on building construction metaphor publication-title: Proceedings of the Advanced Digital Library Conference – year: 2000 ident: bib0125 article-title: Random indexing of text samples for latent semantic analysis publication-title: Proceedings of the 22nd Annual Conference of the Cognitive Science Society – volume: 71 start-page: 176 year: 1983 end-page: 183 ident: bib0020 article-title: Indexing consistency in MEDLINE publication-title: Bull. Med. Libr. Assoc. – reference: U.S. National Library of Medicine. MetaMap Indexing Algorithm. Available from: – volume: 43 start-page: 240 year: 2010 end-page: 256 ident: bib0040 article-title: Reflective random indexing for semi-automatic indexing of the biomedical literature publication-title: J. Biomed. Inform. – volume: 65 start-page: 381 year: 1998 end-page: 405 ident: bib0045 article-title: The structure of a scientific paper publication-title: Philos. Sci. – reference: U.S. National Library of Medicine. Medical Text Indexer (MTI). Bethesda, MD. Available from: – reference: , (updated March 16, cited 6.11.08) (2004). – start-page: 467 year: 2002 end-page: 474 ident: bib0070 article-title: Extracting reputation in multi agent systems by means of social network topology publication-title: Proceedings of the 1st International Joint Conference on Autonomous Agents and MultiAgent Systems (AAMAS) – reference: , (updated November 27, cited 30.03.07) (2006). – year: 1998 ident: bib0050 article-title: Comprehension: A Paradigm for Cognition – reference: , (updated March 16, cited 29.12.09) (2004). – reference: , (cited 21.12.09) (2009). – start-page: 76 year: 2004 end-page: 83 ident: bib0060 article-title: Abstraction summarization for managing the biomedical research literature publication-title: Proc. HLT NAACL Workshop on Computational Lexical Semantics – reference: U.S. National Library of Medicine. PubMed Related Citations algorithm. Available from: – year: 2009 ident: bib0185 publication-title: The Evolution of MetaMap, a Concept Search Program for Biomedical Text. AMIA 2009 November 16 – year: 2005 ident: bib0140 article-title: An introduction to random indexing publication-title: Methods and Applications of Semantic Indexing Workshop at the 7th International Conference on Terminology and Knowledge Engineering – year: 2006 ident: bib0160 article-title: Summary of Threshold Studies. U.S. National Library of Medicine – volume: 107 start-page: 287 year: 2004 ident: bib0175 article-title: Application of a medical text indexer to an online dermatology atlas publication-title: Stud. Health Technol. Inform. – reference: , (updated February 23, cited 18.04.07) (2007). – reference: , (15.03.05) (1998). – start-page: 271 year: 2005 end-page: 275 ident: bib0025 article-title: Semi-automatic indexing of full text biomedical articles publication-title: AMIA Annu. Symp. Proc. – volume: 40 start-page: 288 year: 2007 end-page: 299 ident: bib0120 article-title: Measures of semantic similarity and relatedness in the biomedical domain publication-title: J. Biomed. Inform. – reference: U.S. National Library of Medicine. Trigram Algorithm. Available from: – reference: , (updated March 16, cited 06.11.08) (2004). – volume: 43 start-page: 240 year: 2010 end-page: 256 ident: bib0130 article-title: Reflective random indexing and indirect inference: a scalable method for discovery of implicit connections publication-title: J. Biomed. Inform. – volume: 42 start-page: 814 year: 2008 end-page: 823 ident: bib0005 article-title: A recent advance in the automatic indexing of the biomedical literature publication-title: J. Biomed. Inform. – year: 2004 ident: bib0075 article-title: TextRank: Bringing Order into Texts publication-title: Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP 2004) – reference: U.S. National Library of Medicine. Online services reference manual. Bethesda, MD (1982). – reference: U.S. National Library of Medicine. Medical Subject Heading (MESH) Fact Sheet. (Web page) Bethesda, MD: U.S. National Library of Medicine. Available from: – reference: U.S. National Library of Medicine. MeSH history. (Web page) Bethesda, MD: National Library of Medicine. Available from: – volume: 42 start-page: 390 year: 2009 end-page: 405 ident: bib0115 article-title: Empirical distributional semantics: methods and biomedical applications publication-title: J. Biomed. Inform. – year: 2007 ident: bib0170 article-title: Multiple approaches to fine-grained indexing of the biomedical literature publication-title: Pac. Symp. Biocomput. – year: 2004 ident: 10.1016/j.ijmedinf.2011.02.008_bib0075 article-title: TextRank: Bringing Order into Texts – ident: 10.1016/j.ijmedinf.2011.02.008_bib0080 – ident: 10.1016/j.ijmedinf.2011.02.008_bib0090 – year: 1994 ident: 10.1016/j.ijmedinf.2011.02.008_bib0135 article-title: Context vectors; general purpose approximate meaning representations self-organized from raw data – start-page: 467 year: 2002 ident: 10.1016/j.ijmedinf.2011.02.008_bib0070 article-title: Extracting reputation in multi agent systems by means of social network topology – year: 1998 ident: 10.1016/j.ijmedinf.2011.02.008_bib0050 – volume: 13 start-page: 96 issue: 1 year: 2006 ident: 10.1016/j.ijmedinf.2011.02.008_bib0065 article-title: Using citation data to improve retrieval from MEDLINE publication-title: J. Am. Med. Inform. Assoc. doi: 10.1197/jamia.M1909 – ident: 10.1016/j.ijmedinf.2011.02.008_bib0110 – volume: 107 start-page: 287 issue: Pt 1 year: 2004 ident: 10.1016/j.ijmedinf.2011.02.008_bib0175 article-title: Application of a medical text indexer to an online dermatology atlas publication-title: Stud. Health Technol. Inform. – ident: 10.1016/j.ijmedinf.2011.02.008_bib0095 – ident: 10.1016/j.ijmedinf.2011.02.008_bib0105 – volume: 65 start-page: 381 issue: 3 year: 1998 ident: 10.1016/j.ijmedinf.2011.02.008_bib0045 article-title: The structure of a scientific paper publication-title: Philos. Sci. doi: 10.1086/392651 – year: 2005 ident: 10.1016/j.ijmedinf.2011.02.008_bib0140 article-title: An introduction to random indexing – volume: 71 start-page: 176 issue: 2 year: 1983 ident: 10.1016/j.ijmedinf.2011.02.008_bib0020 article-title: Indexing consistency in MEDLINE publication-title: Bull. Med. Libr. Assoc. – volume: 107 start-page: 268 issue: Pt 1 year: 2004 ident: 10.1016/j.ijmedinf.2011.02.008_bib0145 article-title: The NLM Indexing Initiative's Medical Text Indexer publication-title: Stud. Health Technol. Inform. – year: 2009 ident: 10.1016/j.ijmedinf.2011.02.008_bib0185 – ident: 10.1016/j.ijmedinf.2011.02.008_bib0015 – year: 2006 ident: 10.1016/j.ijmedinf.2011.02.008_bib0160 – volume: 40 start-page: 288 year: 2007 ident: 10.1016/j.ijmedinf.2011.02.008_bib0120 article-title: Measures of semantic similarity and relatedness in the biomedical domain publication-title: J. Biomed. Inform. doi: 10.1016/j.jbi.2006.06.004 – start-page: 271 year: 2005 ident: 10.1016/j.ijmedinf.2011.02.008_bib0025 article-title: Semi-automatic indexing of full text biomedical articles – ident: 10.1016/j.ijmedinf.2011.02.008_bib0010 – ident: 10.1016/j.ijmedinf.2011.02.008_bib0085 – year: 1998 ident: 10.1016/j.ijmedinf.2011.02.008_bib0165 article-title: KeyGraph: automatic indexing by co-occurrence graph based on building construction metaphor – year: 2006 ident: 10.1016/j.ijmedinf.2011.02.008_bib0155 article-title: Nonparametrics – year: 2007 ident: 10.1016/j.ijmedinf.2011.02.008_bib0170 article-title: Multiple approaches to fine-grained indexing of the biomedical literature – ident: 10.1016/j.ijmedinf.2011.02.008_bib0035 – volume: 43 start-page: 240 issue: 2 year: 2010 ident: 10.1016/j.ijmedinf.2011.02.008_bib0040 article-title: Reflective random indexing for semi-automatic indexing of the biomedical literature publication-title: J. Biomed. Inform. – volume: 42 start-page: 814 issue: 5 year: 2008 ident: 10.1016/j.ijmedinf.2011.02.008_bib0005 article-title: A recent advance in the automatic indexing of the biomedical literature publication-title: J. Biomed. Inform. doi: 10.1016/j.jbi.2008.12.007 – ident: 10.1016/j.ijmedinf.2011.02.008_bib0150 – volume: 42 start-page: 390 issue: 2 year: 2009 ident: 10.1016/j.ijmedinf.2011.02.008_bib0115 article-title: Empirical distributional semantics: methods and biomedical applications publication-title: J. Biomed. Inform. doi: 10.1016/j.jbi.2009.02.002 – start-page: 319 year: 2001 ident: 10.1016/j.ijmedinf.2011.02.008_bib0030 article-title: Automatic MeSH term assignment and quality assessment – ident: 10.1016/j.ijmedinf.2011.02.008_bib0180 – year: 2000 ident: 10.1016/j.ijmedinf.2011.02.008_bib0125 article-title: Random indexing of text samples for latent semantic analysis – start-page: 76 year: 2004 ident: 10.1016/j.ijmedinf.2011.02.008_bib0060 article-title: Abstraction summarization for managing the biomedical research literature – year: 1976 ident: 10.1016/j.ijmedinf.2011.02.008_bib0055 – ident: 10.1016/j.ijmedinf.2011.02.008_bib0100 – volume: 43 start-page: 240 issue: 2 year: 2010 ident: 10.1016/j.ijmedinf.2011.02.008_bib0130 article-title: Reflective random indexing and indirect inference: a scalable method for discovery of implicit connections publication-title: J. Biomed. Inform. doi: 10.1016/j.jbi.2009.09.003 |
| SSID | ssj0017054 |
| Score | 1.9982727 |
| Snippet | ► We define, implement and evaluate MEDRank. ► MEDRank is a graph-based algorithm that identifies important concepts in text. ► MEDRank improves retrieval of... Highlights ► We define, implement and evaluate MEDRank. ► MEDRank is a graph-based algorithm that identifies important concepts in text. ► MEDRank improves... As the volume of biomedical text increases exponentially, automatic indexing becomes increasingly important. However, existing approaches do not distinguish... |
| SourceID | pubmedcentral proquest pubmed crossref elsevier |
| SourceType | Open Access Repository Aggregation Database Index Database Enrichment Source Publisher |
| StartPage | 431 |
| SubjectTerms | Abstracting and Indexing - methods Abstracting and indexing as topic Algorithms Artificial Intelligence Automatic data processing Digital libraries Electronic Data Processing Humans Information Storage and Retrieval Internal Medicine Medical informatics Medical Subject Headings MEDLINE National Library of Medicine (U.S.) Natural language processing Other PubMed Software United States |
| Title | MEDRank: Using graph-based concept ranking to index biomedical texts |
| URI | https://www.clinicalkey.com/#!/content/1-s2.0-S1386505611000542 https://www.clinicalkey.es/playcontent/1-s2.0-S1386505611000542 https://dx.doi.org/10.1016/j.ijmedinf.2011.02.008 https://www.ncbi.nlm.nih.gov/pubmed/21439897 https://www.proquest.com/docview/864780136 https://www.proquest.com/docview/874190820 https://pubmed.ncbi.nlm.nih.gov/PMC3090689 |
| Volume | 80 |
| WOSCitedRecordID | wos000290017100006&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: 1872-8243 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0017054 issn: 1386-5056 databaseCode: AIEXJ dateStart: 19970301 isFulltext: true titleUrlDefault: https://www.sciencedirect.com providerName: Elsevier |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1bb9MwFLbKhhAviDvlMuWBtyojFye2eZtY0YbUCa1F6pvlpM7WdqRT0lXjl_H3OCd20pRyGZN4iao0blyfL-eWcz4T8pYyJuIo1W6aZpFLwUV2VZwErtJhynkcMZql1WYT7OSEj8fic6fzve6FWV2wPOfX1-Lyv4oazoGwsXX2H8Td_CicgM8gdDiC2OF4I8EP-oenKp9jqG_qASpOahfNFXawVV2KPdyqvWqUWvQqvsSeacOvJIa1IGXbad3MGra4JuoRln21YnwGj3V4rvLctpRhrwzmrpt8wxH4m_MF6CeTuy_OdO90v3kTUjeLjAq9WjR1w6DdCoVEHcrqyNV03piT4286PzNl4oP93rCYFuprO5fht2quTIKtbrKxk2tVNqFuDnnsosPWVt5mGygL0rYmpta4GKNODbvWlr0wqYvZ_nSGK5bXnK5I4srXFrKpWxziHHAKSLQHzi7Y_t2ARQLU6e7BcX_8qXmBxbzI7Kds59xqTv_13X7nF23HPT-X77b8odFD8sAGMs6BAeAj0tH5Y3JvYEs1npBDi8P3ToVCp4VCx6LQsSh0lgunQqGzRqFTofAp-fKxP_pw5NotO9w0iunS5YH2teY80T4NvEk0oRAW0TAJuadZlKYJj3USCl_5ScI0-P7g7_si89lE81hlQfiM7ABA9QviTMCWZKmi4EAFNEk8FWc-TbSGCCNCYr4uieoFk6nls8dtVS5kXbg4k_VCS1xo6QUSFrpL3jXjLg2jy19HsFoesu5XBgsrAUS3G6lL-6iW0pclXCm3gNUlohlpfWHj497ork4NGgnGAt8AqlwvrkrJsbMcWRr_cAmEGALjgi55bmDWrFIAsZXggsG_2gBgcwFS1W9-k0_PK8r60BNezMXLW6_kK3J_rTBek51lcaXfkLvpajktiz1yh435nn0GfwAEOBHW |
| 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=MEDRank%3A+Using+graph-based+concept+ranking+to+index+biomedical+texts&rft.jtitle=International+journal+of+medical+informatics+%28Shannon%2C+Ireland%29&rft.au=Herskovic%2C+Jorge+R.&rft.au=Cohen%2C+Trevor&rft.au=Subramanian%2C+Devika&rft.au=Iyengar%2C+M.+Sriram&rft.date=2011-06-01&rft.pub=Elsevier+Ireland+Ltd&rft.issn=1386-5056&rft.volume=80&rft.issue=6&rft.spage=431&rft.epage=441&rft_id=info:doi/10.1016%2Fj.ijmedinf.2011.02.008&rft.externalDocID=S1386505611000542 |
| thumbnail_m | http://cvtisr.summon.serialssolutions.com/2.0.0/image/custom?url=https%3A%2F%2Fcdn.clinicalkey.com%2Fck-thumbnails%2F13865056%2FS1386505611X00060%2Fcov150h.gif |