eNote - Electronic Notes in Medicine

This project seeks to capture and structure narrative in the online medical record, in order to improve entry time, completeness, information content and consistency of clinical documentation.

Narrative information is vital to health care, because it enables physicians to synthesize the raw facts and provide a context and interpretation for them. Electronic medical record systems contain a wealth of clinical data, but typically lack the clinical narrative found in paper records, e.g., the patient history and progress notes. Numerous barriers prevent the timely acquisition of narrative data, and most computer systems are unable to use such information productively. Current approaches offer a tradeoff: capture of rich clinical data that lacks structure (using transcription services or speech technology), versus entry of structured data that lacks flexibility and expressiveness (using template systems). Natural language processing can integrate these approaches by allowing physicians full freedom of expression while producing structured documents that preserve the richness and enable further computer processing.

Objectives

1. Maintain the continuity of the medical record: A lengthy medical record requires significant time to review and digest. Many facts from past narratives remain true in the present or persist with minor changes. By automatically bringing these facts forward into the current narrative, the system can reduce the time to enter the document, and improve the completeness of documentation by maintaining continuity of what is known about a patient.
2. Integrate the medical record: Electronic medical records contain a vast amount of data. However, most of these data are raw facts. By helping the physician to connect, interpret and summarize these facts, the system can improve the usefulness of the information in the record, and reduce the time to enter documents by performing some syntheses automatically.
3. Harmonize the medical record: The multidisciplinary nature of health care creates the potential for the differing perspectives and interpretations in the medical record, and even contradictions. By bringing together information from different kinds of documents, the system can keep clinicians informed and help reduce inconsistencies.

Components

• Knowledge base defining hundreds of kinds of clinical documents and their characteristics. This is being standardized in conjunction with Logical Observation Identifiers Names and Codes (LOINC)
• XML database storing documents represented using the Clinical Document Architecture (CDA) of Health Level 7 (HL7)
• Novel user interface for browsing clinical documents and rapidly creating new ones.
• Automated markup of clinical information using natural language processing to interpret clinical narratives.
• Temporal analysis module, which determines how medical events in narrative are organized in time.
• Automatic detection of medical errors in narrative (DETER-MINE) using data mining techniques

Personnel

Meetings

Research: Mondays 10:30-11:30am, School of Nursing, CEBP Research Office (ground floor) Technical: Fridays 9:30-10:30am, Presbyterian Hosptial, 9th Floor, East Conference Room

Discussion List

You can post messages to enote atsign dbmi. Please send mail to Professor Johnson if you would like to be added to this list.

Publications

• Hyun S, Bakken S, Friedman C, Johnson SB.Natural language processing challenges in HIV/AIDS clinic notes, AMIA Annu Symp 2003
• Hyun S, Bakken S, Friedman C, Johnson SB.Natural language processing challenges in HIV/AIDS clinic notes, Medinfo 2004
• Bakken S, Hyun S, Friedman C, Johnson SB. A comparison of semantic categories of the ISO Reference Terminology Models for Nursing and the MedLEE natural language processing system. MedInfo 2004
• George Hripcsak, Li Zhou, Simon Parsons, Amar K. Das, Stephen B. Johnson Modeling Electronic Discharge Summaries as a Simple Temporal Constraint Satisfaction Problem. JAMIA 2004.
• Li Zhou, George Hripcsak, Simon Parsons, Amar K. Das, Stephen B. Johnson. Reasoning about Time in Electronic Discharge Summaries Using Temporal Constraint Satisfaction Techniques. Intelligent Data Analysis In Medicine And Pharmacology 2004.
• Bakken S, Hyun S, Friedman C, Johnson SB. ISO reference terminology models for nursing: applicability for natural language processing of nursing narratives, Medinfo 2004
• Haas J, Bright TJ, Bakken S, Stetson P, Johnson SB. Clinician perceptions of the usability of eNote. Proc AMIA Symp. 2005:973.
• Stetson PD, Lee PG, Bakken S, Johnson SB, Haas J, Chang JC, Shea S. Electronic Clinical Documentation to Support Patient Safety. Society of Hospital Medicine, Innovations Track, Chicago 2005.
• Bakken S, Hyun S, Melton-Meaux G, Schlegel CS, Mendonca E, Johnson SB. Evaluation of the HL7-LOINC Document Ontology. JAMIA 2005.
• Stetson PD, Keselman A, Rappaport D, Van Vleck T, Cooper M, Boyer A, Hripcsak G. Electronic Discharge Summaries. Proceedings of AIMA 2005:1121-5.
• Zhou L, Melton GB, Parsons S, Hripcsak G. A temporal constraint structure for extracting temporal information from clinical narrative. Journal of Biomedical Informatics 2005(in press).
• Shapiro JS, Bakken S, Hyun S, Melton GB, Schlegel C, Johnson SB, Document Ontology: Supporting Narrative Documents in Electronic Health Records. Proc AMIA Symp 2005:684-8.
• Bakken S, Hyun S, Friedman C, Johnson SB. ISO reference terminology models for nursing: applicability for natural language processing of nursing narratives, IJMI 2005 Aug.
• Hyun S, Bakken S, Johnson SB. Markup of Temporal Information in Electronic Health Records, International Congress in Nursing Informatics Conference 2006 (submitted).
• Hyun S, Ventura R, Johnson SB, Bakken S. Is the Health Level 7/LOINC Document Ontology Adequate for Representing Nursing Documents?, International Congress in Nursing Informatics Conference 2006 (submitted).
• Bright TJ, Bakken S, Johnson SB. A heuristic evaluation of eNote. JAMIA 2006 (in preparation).