Download PDF
Methods Inf Med 2003; 42(01): 51-60
DOI: 10.1055/s-0038-1634209
Original article
Schattauer GmbH

Text Generation in Clinical Medicine – a Review

D. Hüske-Kraus
1   Heart and Diabetes Centre, Northrhine-Westphalia, Bad Oeynhausen, Germany
› Author Affiliations
Further Information

Publication History

Publication Date:
07 February 2018 (online)

    Permissions and Reprints

Summary

Objectives: This article aims at an analysis of ways of producing documents (such as findings or referral letters) in clinical medicine. Special emphasis is given to the question of whether the field of “Natural Language Generation” (NLG) can provide new approaches to ameliorate the current situation.

Methods: In order to assess the currently used techniques in text production, an analysis of commercially available systems was performed in addition to an extensive review of the literature. The sketch of current NLG approaches is also based on a literature review.

To estimate the applicability of several techniques to clinical documents, a typology of documents in clinical medicine was developed, based on rhetorical structure theory, speech act theory and certain recurrent linguistic phenomena exposed in the said documents. Results: Current ways of producing text for documents in medicine are less than optimal in several respects. The field of NLG draws on the idea of generating text from a conceptual representation of not only certain facts, but also knowledge about how to express them via (written) language.

Unfortunately, NLG does not yet offer “ready-to-run” solutions for the automatic production of most of the document types in the given typology.

It seems, however, highly plausible that the demands of medical informatics for these kinds of systems will be satisfiable as NLG matures.

Conclusions: NLG offers a promising way of generating text for clinical documents, a problem of enormous economical importance. The medical informatics community should therefore commit itself to the idea of NLG in medicine.

Keywords

Natural language generation - clinical documents - document typology
 

  • References

  • 1 Hons J. Der elektronische Arztbrief ist ein Stück näher gerückt. Ärztezeitung; 2001. 5/2001 (3); Sect. 21.
    Google Scholar
  • 2 Wagner JC, Rogers JE, Baud RH, Scherrer J-R. Natural language generation of surgical procedures. Int J Med Inform 1999; 53: 175-92.
    CrossrefPubMedGoogle Scholar
  • 3 Hahn U, Romacker M, Schulz S. Discourse structure in medical reports - Watch out! The generation of referentially coherent and valid text knowledge bases in the medSyndikate system. Int J Med Inform 1999; 53: 1-28.
    CrossrefPubMedGoogle Scholar
  • 4 Cawsey AJ, Webber BL, Jones RB. Natural language generation in health care. J Am Med Inform Assoc 1997; 4 (Suppl. 06) 473-82.
    CrossrefPubMedGoogle Scholar
  • 5 Olympus. ENDOBASE; 2001 www.olympus-europa.nl/endo_int/products/endobase/features.stm
    PubMed
  • 6 Medistar. Medistar; 2001 www.medistar.de
    PubMed
  • 7 GWI. ORBIS; 2001 www.orbis-net.de
    PubMed
  • 8 DragonSystems, Dragon Naturally Speaking Medical V6.0. 2002. www.voicerecognition.com dragon.
    PubMed
  • 9 IBM, Via Voice for Windows, Release 9. 2002 www.speechtechnology.com/ibm/med_vocab.html.
    PubMed
  • 10 Lussier YA, Maksud M, Desruisseaux B, Yale P-P, St-Arneault R. Pure MD. In: A computerized patient record software for direct data entry by physicians using a keyboard-free pen-based portable computer. Frisse ME. ed. SCAMC 92; 1992;. Baltimore: McGraw-Hill; 1992. p. 261-3.
    Google Scholar
  • 11 Heathfield HA, Kirby J, Hardiker N. Data entry in computer-based care planning. In: Medical Informatics in Europe 94 (MIE 94); 1994;. 1994 p. 186-9.
    PubMedGoogle Scholar
  • 12 Ginneken AMv. Structured data entry in ORCA: the strengths of the two models combined. In: 20th Symposium on Computer Applications Med Care (SCAMC 96);. 1996. Hanley & Belfus, Philadelphia;: 1996. p. 767-801.
    Google Scholar
  • 13 Colburn C. Structured text-documentation meets technology. J AHIMA. 1997 2.
    PubMedGoogle Scholar
  • 14 Kuhn K, Zemmler T, Reichert M, Heinlein C, Roesner D. Structured data collection and knowledge-based user guidance for abdominal ultrasound reporting. In: Proceedings Ann Symposium on Comput Applied Med Care;. 1993 p. 311-5.
    PubMedGoogle Scholar
  • 15 Moorman PW, Ginneken AMv, Lei Jvd, Bemmel JHv. A model for structured data entry based on explicit descriptional knowledge. Methods Inf Med 1994; 33: 454-63.
    Article in Thieme ConnectPubMedGoogle Scholar
  • 16 Yamazaki S, Satomura Y. Standard method for describing an electronic patient record template: Application of XML to share domain knowledge. Methods Inf Med 2000; 39: 50-5.
    Article in Thieme ConnectPubMedGoogle Scholar
  • 17 Hohnloser JH, Engelmeier T. Dateneingabe durch Ärzte in elektronische Patientenakten: Vergleich zweier Methoden im 5-jährigen klinischen Routinebetrieb. In: GMDS; 96; 1996
    Google Scholar
  • 18 Kraus D. Suregen2 – a model based generator for surgical reports. In: MIE 2000. IOS Press; 2000
    Google Scholar
  • 19 DeFriece RJ. MedIO: A program for intelligent clinical data entry. In: 19th Annual Symposium on Computer Applications in Medical Care (SCAMC 95). Gardner RM. ed. New Orleans, Louisiana, USA: Hanley & Belfus; 1995. p 1005.
    Google Scholar
  • 20 Zacher M. Computergestützte Befunddokumentation in der Arthrosonographie (M. D.). Giessen: Justus-Liebig-Universität Giessen; 1994
    Google Scholar
  • 21 Veigel B. OrthoStar; 2001 www.medistar.de/sites/fachaerzte/orthostar.html
    PubMed
  • 22 AGP. Phoenix; 2001 www.phoenix.ch
    PubMed
  • 23 Kahn CE. A generalized language for platform-independent structured reporting. Methods Inf Med 1997; 36: 163-71.
    Article in Thieme ConnectPubMedGoogle Scholar
  • 24 McCray AT, Hole WT. The scope and structure of the first version of the UMLS semantic network. In: Fourteenth Annual Symposium on Computer Applications in MedicalCare (SCAMC-90). Miller RA. ed. Washington DC: IEEE Computer Society Press; 1990. p. 126-33.
    Google Scholar
  • 25 McCray AT, Nelson SJ. The representation of meaning in the UMLS. Methods Inf Med 1995; 34 1-2 193-201.
    Article in Thieme ConnectPubMedGoogle Scholar
  • 26 Mönnich G, Wetter T. Requirements for speech recognition to support medical documentation. Methods Inf Med 2000; 39 (Suppl. 01) 63-9.
    Article in Thieme ConnectPubMedGoogle Scholar
  • 27 Nugues N, Godéreaux C, Guedj POE, Cazenave F. Question answering in an oral dialogue system. In: Annual International Conference of the IEEE Engineering Medicine Biology Society; 1993. p. 590-1.
    Google Scholar
  • 28 Pereira FC, Warren DH. Definite clause grammars for natural language analysis. Artificial Intelligence. 1980 13.
    PubMedGoogle Scholar
  • 29 Mann WC, Thompson SA. Rhetorical structure theory: A theory of text organisation. Technical Report. Marina del Rey CA, USA: USC/Information Sciences Institute; 1987. Report No.: RS-87-190.
    Google Scholar
  • 30 Moore JD, Pollack ME. A problem for RST: The need for multi-level discourse analysis. Computational Linguistics 1992; 18 (Suppl. 04) 537-44.
    PubMedGoogle Scholar
  • 31 Rosner D, Stede M. Customizing RST for the automatic production of technical manuals. In: Sixth International Workshop on Natural Language Generation; 1992;. Duisburg: Springer Verlag, Heidelberg; 1992. p. 199-215.
    Google Scholar
  • 32 Dalianis H, Johannesson P. Explaining conceptual models - using toulmin’s argumentation model and RST. In: Third International Workshop on the language Action Perspective on Communication Modelling (LAP 98); 1998;. Stockholm, Schweden; 1998. p. 131-40.
    Google Scholar
  • 33 Cole RA, Mariani J, Uszkoreit H, Zaenen A, Zue V. Survey of the State of the Art in human language technology. 1996 www.cslu.cse.ogi.edu/HLTsurvey/HLTSurvey.html
    PubMedGoogle Scholar
  • 34 Shaw J. Conciseness through aggregation in text generation. In: 33rd ACL (Student Session); 1995;. 1995. p. 329-31.
    Google Scholar
  • 35 Dalianis H. Concise natural language generation from formal specifications. PhD-Thesis. Stockholm: Dept Comput Systems Science-Royal Institute of Technology, Stockholm University; 1996. 6/1996. Report No.: 96-008.
    Google Scholar
  • 36 Dalianis H, Hovy E. On lexical aggregation and ordering. In: Sixth European Workshop on Natural Language Generation; 1997;. Duisburg: 1997. p. 17-27.
    Google Scholar
  • 37 Shaw J. Segregatory coordination and ellipsis in text generation. In: 36th Annual Meeting of the Association for Computational Linguistics and the 17th International Conference on Computational linguistics (COLING-ACL 98); 1998;. 1998 p. 1220-6.
    PubMedGoogle Scholar
  • 38 Reape M, Mellish C. Just what is aggregation anyway?. In: 7th European Workshop on Natural Language Generation (EWNLG 99); 1999;. Toulouse; 1999. p. 20-9.
    Google Scholar
  • 39 Shaw J. Clause Aggregation using linguistic knowledge. In: 9th International Workshop on Natural Language Generation; 1998;. Niagara-on-the-Lake, Canada: 1998
    Google Scholar
  • 40 Grosz B. The representation and use of focus in dialogue unterstanding. Technical Report.
    PubMedGoogle Scholar
  • 41 Kamp H. A theory of truth and semantic interpretation. In: Truth, Interpretation and Information. Dordrecht: Foris Publications; 1981
    Google Scholar
  • 42 Grice HP. Logic and conversation. In: Speech acts. Cole P, Morgan JL. eds. New York, USA: Academic Press; 1975. p. 41-58.
    Google Scholar
  • 43 Searle JR. Speech Acts: An essay in the philosophy of language. Cambridge: Cambridge University Press; 1969
    Google Scholar
  • 44 Austin JL. How to do things with words. Oxford: Clarendon Press; 1962
    Google Scholar
  • 45 Akmajian A, Demers RA, Farmer AK, Harnish RM. Linguistics, an introduction to language and communication. Cambridge MA, USA: MIT Press; 1990
    Google Scholar
  • 46 Stent AJ. Aspects of natural language generation. Technical Report. Rochester: University of Rochester; 1998. 11/1998. Report No.: 701
    Google Scholar
  • 47 Bateman J, Rondhuis KJ. Coherence relations: analysis and specification. project deliverable. Darmstatd: GMD-IPSI; 1994. 6
    Google Scholar
  • 48 Bateman J, Zock M. Updated list of generation systems;. 2001 www.fb10.uni-bremen.de/anglistik/langpro/NLG-table/NLG-table-root.htm
    PubMedGoogle Scholar
  • 49 Goldberg E, Driedger N, Kittredge R. Using natural language processing to produce weather forecasts. IEEE Expert 1994; 9 (Suppl. 02) 45-53.
    PubMedGoogle Scholar
  • 50 Reiter E, Dale R. Building applied natural language generation systems. J Natural Language Engineering 1997; 3 (Suppl. 01) 57-87.
    PubMedGoogle Scholar
  • 51 Paiva DS. A survey of applied natural language generation systems. Technical Report. Brighton: Information Technology Research Institute; 1998. 7/1998. Report No.: ITRI- 98-03.
    Google Scholar
  • 52 Cahill L, Evans R, Mellish C, Reape M, Scott D. Towards a reference architecture for natural language generation systems – The RAGS project. Technical Report. Brighton: Inform Technology Res Institute; 2000. 8/1999.
    Google Scholar
  • 53 Elhadad M. The FUF Functional unifier: user’s manual (version 3.0). Technical Report. New York: Columbia University, Dept. of Computer Science; 1990. Report No.: CUCS-012-91.
    Google Scholar
  • 54 Elhadad M, Robin J. An overview of SURGE: a reusable comprehensive syntactic realization component. Technical Report. Beer Sheva, Israel: Dept. Comput Science, Ben Gurion University; 1996. 4/1996. Report No.: 96-03.
    Google Scholar
  • 55 Bateman J. Enabling technology for multilingual natural language generation: The KPML development environment. Natural Language Engineering 1997; 3: 15-55.
    CrossrefPubMedGoogle Scholar
  • 56 Lavoie B, Rambow O. a fast and portable realizer for text generation systems. In: Conference on Applied Natural Language Processing (ANLP’97); 1997 July 1997;. Washington, DC: Melamed, I. D; 1997. p. 265-8.
    Google Scholar
  • 57 CoGenTex. Tools; 2001 www.cogentex.com technology/tools.html
    PubMed
  • 58 Wilkinson J. Aggregation in natural language generation: Another look (M. Sc.). Waterloo: University of Waterloo, Canada; 1995
    Google Scholar
  • 59 Rösner D, Stede M. Generating multilingual documents from a knowledge base: The TECHDOC Project. In: 15th International Conference on Computational Linguistics (COLING 94); 1994;. Kyoto: 1994
    Google Scholar
  • 60 Wagner JC, Rassinoux A-M, Baud RH, Scherrer J-R. Generating noun phrases from a medical knowledge representation. In: MIE 94; 1994;. Lisbon, Portugal: Springer; 1994. p. 218-23.
    Google Scholar
  • 61 Wagner JC, Richardson WD, Michel P-A, Baud RH, Rector AL, Scherrer J-R. Multilingual natural language generation as part of a medical terminology server. In: MEDINFO 95. Greenes RA, Peterson HE, Protti DJ. eds. Coach Publishing, Vancouver; 1995. p. 100-4.
    Google Scholar
  • 62 Rector A, Zanstra PE, Solomon DW, Rogers JE, Baud R, Ceusters W. et al. Reconciling users needs and formal requirements: issues in developing a reusable ontology for medicine. IEEE Transactions Inform Biomed 1998; 2 (Suppl. 04) 239-42.
    PubMedGoogle Scholar
  • 63 Dale R. Generating referring expressions: Constructing Descriptions in a domain of objects and processes. Cambridge MA, USA: MIT Press; 1992
    Google Scholar
  • 64 Shaw J, McKeown K. Generating referring quantified expressions. In: 1st International Natural Language Generation Conference; 2000;. Mitzpe Ramon, Israel: 2000: 100-7.
    Google Scholar
  • 65 Eugenio Bd, Moore JD, Rosé CP. Generating ‘distributed’ referring expressions: an initial report. In: Eighth Int Workshop Natural Language Generation (INLG 96); 1996;. Herstmonceux, France: 1996
    Google Scholar
  • 66 Deemter Kv. Generating vague descriptions. In: Int Natural Language Generation Conference (INLG-2000); 2000; Mitzpe-Ramon, Israel: 2000. p. 179-85.
    Google Scholar
  • 67 Yeh C-L, Mellish C. An empirical study on the generation of anaphora in chinese. Computational Linguistics 1997; 23 (Suppl. 01) 169-90.
    PubMedGoogle Scholar
  • 68 Jameson A, Wahlster W. User modelling in anaphora generation: ellipsis and definite description. In: Europ Conference Artificial intelligence (ECAI 82); 1982; Orsay, France: 1982. p. 222-7.
    Google Scholar
  • 69 Dale R, Reiter E. Computational interpretations of the Gricean Maxims in the generation of referring expressions. Cognitive Science 1995; 19: 233-63.
    CrossrefPubMedGoogle Scholar
  • 70 Dale R. Generating one-anaphoric expressions:Where does the decision lie?. In: PACLING-II; 1995;. Brisbane, Australia: 1995. p. 49-58.
    Google Scholar
  • 71 Partee BH. Nominal and temporal anaphora. Linguistics Philosophy 1984; 7: 243-86.
    PubMedGoogle Scholar
  • 72 Geldof S, Velde Wvd. An architecture for template based hypertext generation. In: 6th Europ Workshop Natural Language Generation (EWNLG 97). Hoeppner W. ed. Duisburg; 1997. p. 28-37.
    Google Scholar
  • 73 DiMarco C, Hirst G, Hovy E. Generation by selection and repair as a method for adapting text for the individual reader. In: Workshop on flexible Hypertext (8th ACM International Hypertext Conference; 1997 May 1997; Southampton: 1997
    Google Scholar
  • 74 Wahlster W, Andre E, Finkler W, Profitlich H, Rist T. Plan-based integration of natural language and graphics generation. Artificial Intelligence 1993; 63: 387-427.
    CrossrefPubMedGoogle Scholar
  • 75 Andre E, Finkler W, Graf W, Rist T, Schauder A, Wahlster W. The automatic synthesis of multimodal presentations. In: Maybury M. ed. Intelligent Multimedia Interfaces. Menlo Park: MIT Press: 1993. p. 75-93.
    Google Scholar
  • 76 Reiter E, Osman L. Tailored patient information: some issues and questions. In: From research to commercial applications: Making NLP technology work in practice (ACL-1997 workshop). Madrid: 1997. p. 29-34.
    Google Scholar
  • 77 Hirst G, DiMarco C, Hovy E, Parsons K. Authoring and generating health-education documents that are tailored to the needs of the individual patient. In: 6th Int Conference User Modeling, 1997. Sardinia, Italy;: 1997. p. 107-18.
    Google Scholar
  • 78 Elhadad N, McKeown KR. Towards generating patient specific summaries of medical articles. 2001
    PubMedGoogle Scholar
  • 79 McKeown K, Pan S, Shaw J, Jordan D, Allen B. Language generation for multimedia health-care briefings. In: Fifth ACL Conference on Applied Natural Language Processing (ANLP-97); 1997 31.3-4.3. Washington D.C.;: 1997. p. -277 -82.
    Google Scholar
  • 80 Li P-Y, Evans M, Hier D. Generating medical case reports with the linguistic string parser. In: Fifth National Conference on Artificial Intelligence (AAAI 86). August 11-15, 1986; Philadelphia, PA: Morgan Kaufmann; 1986. p. 1069-73.
    Google Scholar
  • 81 Giere W. BAIK - Befunddokumentation und Arztbriefschreibung im Krankenhaus. Taunusstein: Media; 1986
    Google Scholar
  • 82 McDonald CJ, Tierney WM, Blevins L. The benefits of automated medical record systems for ambulatory care. In: Implementing Health Care Inform Systems. Orthner HF, Blum BI. eds. New York: Springer; 1988
    Google Scholar
  • 83 Kraus D, Miche E. Arztbrief – generating medical reports in a multimedia environment. In: Medical Informatics Europe (MIE 91). Adlassnig KP, Grabner G, Bengtsson S, Hansen R. eds. 1991; Wien: Springer; 1991. p. 900-4.
    Google Scholar
  • 84 Bernauer J, Gumrich K, Kutz S, Linder P, Pretschner P. An interactive report generator for bone scan studies. In: Proceedings Ann Symposium Computers Applied Medical Care (SCAMC 92). Clayton PD. ed. McGraw-Hill (IEEE Computer Society Press); 1992. p. 858-60.
    Google Scholar
  • 85 Bernauer J. Conceptual graphs as an operational model for descriptive findings. In: SCAMC 92. Hanley & Belfus; 1992. p. 214-7.
    Google Scholar
  • 86 Nowlan W, Kay S, Rector AS, Horan B, Wilson A. PEN&PAD: A multilingual patient care workstation based on a unified representation of the medical record and medical terminology. In: MIE 91; 1991;. Vienna Austria: Springer; 1991. p. 1043-8.
    Google Scholar
  • 87 Bullock JC. Text generation from semantic network based medical records (Master’s thesis). Manchester: University of Manchester; 1994
    Google Scholar
  • 88 Gregory J, Mattison JE, Linde C. Naming notes: transitions form free text to structured entry. Methods Inf Med 1995; 34 1-2 57-67.
    Article in Thieme ConnectPubMedGoogle Scholar
  • 89 Birkmann C, Diedrich T, Ingenerf J, Rogers J, Moser W, Engelbrecht R. A formal model of diabetological terminology and its application for data entry. In: Med Inform Europe 97 (MIE ‘97). Pappas C, Maglaveras N, Scherrer J-R. eds. Amsterdam: IOS Press; 1997. p. 426-30.
    Google Scholar
  • 90 Tange HJ, Hasman A, Robbé PFdV, Schouten HC. Medical narratives in electronic medical records. Int J Med Informatics 1997; 46: 7-29.
    CrossrefPubMedGoogle Scholar
  • 91 Reiter E, Dale R. Building natural language generation systems. Cambridge: Cambridge University Press; 2000
    Google Scholar
  • 92 Stede M. Lexical semantics and knowledge representation in multilingual text generation. Kluwer Academic Publishers; 1999
    Google Scholar
  • 93 Ceusters W, Spyns P, Moor GD, Martin W. Syntactic-semantic tagging of medical texts: The Multi-TALE Project. Amsterdam: IOS Press; 1998
    Google Scholar
  • 94 Coch J. Overview of AlethGen. In: 8th Int Workshop natural language generation (INLG 96); 1996;. Herstmonceux, France: 1996. p. 25-8.
    Google Scholar
  • 95 Calder J, Evans R, Mellish C, Reape M. “Free choice” and templates: How to get both at the same time. In: “May I speak freely?” Between templates and free choice in natural language generation. Becker T, Busemann S. eds. Saarbrücken; DFKI Saarbrücken report D-99-01; 1999. p. 19-24.
    Google Scholar
  • 96 Bangalore S, Rambow O, Whittaker S. Evaluation metrics for generation. In: International Natural Language Generation Conference (INLG 2000); 2000;. Mitzpe Ramon, Israel: 2000
    Google Scholar
  • 97 Rector AR. Clinical terminology: Why is it so hard?. Methods Inf Med 1999; 38 4-5 239-52.
    Article in Thieme ConnectPubMedGoogle Scholar
  • 98 Mori AR. Archetypes of situations for clinical information - Terminology;. 2001 www.intratext.com/y/CNR0001.htm
    PubMedGoogle Scholar
  • 99 Rossi-Mori A, Consorti F, Galeazzi E. Standards to support development of terminological systems for healthcare telematics. In: Conference on Natural Language and Medical Concept Representation. WG6 I, ed. Methods Inf Med 1998; 37: 551-63.
    Article in Thieme ConnectPubMedGoogle Scholar
  • 100 Gangemi A, Pisanelli DM, Steve G. Some requirements and experiences in integrating terminological ontologies in medicine. In: KAW 98;. 1998
    PubMedGoogle Scholar
  • 101 Haux R. Health care in the Information Society: What should be the role of medical informatics?. Methods Inf Med 2002; 1: 31-4.
    Article in Thieme ConnectPubMedGoogle Scholar
  • 102 Lei Jvd. Closing the loop between clinical practice, research, and education. The potential of electronic patient records. Methods Inf Med 2002; 41: 51-4.
    Article in Thieme ConnectPubMedGoogle Scholar
  • 103 O’Keefe QEW, Simborg DW, Epstein WV, Warger A. A computerized summary medical record system can provide more information than the standard medical record. JAMA 1985; 254: 1185-92.
    CrossrefPubMedGoogle Scholar
  • 104 Kuhn K, Gaus W, Wechsler JG, Janowitz P, Tudyka J, Kratzer W. et al. Structured reporting of medical findings: evaluation of a system in gastroenterology. Methods Inf Med 1992; 31: 268-74.
    Article in Thieme ConnectPubMedGoogle Scholar
  • 105 Reiter E, Robertson R, Osman L. Lessons from a failure: generating tailored smoking cessation letters. To appear in: Artificial Intelligence 2003 www.csd.abdn.ac.uk/~ereiter/papers/aij03.pdf.
    PubMedGoogle Scholar
  • 106 Sripada SG, Reiter E, Hunter J. Summarising neonatal time series data (accepted for EACL ’03). 2003 www.csd.abdn.ac.uk/~ereiter/transfer/eacl03.doc.
    PubMedGoogle Scholar