Enhancing rule-based text classification of neurosurgical notes using filtered feature weight vectors

Sedigheh Khademi , Pari Delir Haghighi, Frada Burstein, Christopher Palmer

    Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

    Abstract

    Clinicians need to record clinical encounters in written or spoken language, not only for its work-flow naturalness but also for its expressivity, precision, and capacity to convey all required information, which codified structured data is incapable of. Therefore, the structured data which is required for aggregation and analysis must be obtained from clinical text as a later step. Specialised areas of medicine use their own clinical language and clinical coding systems, resulting in unique challenges for the extraction process. Rule-based information extraction techniques have been used effectively in commercial systems and are favoured because they are easily understood and controlled. However, there is promising research into the use of machine learning techniques for extracting information,and this research explores the effectiveness of a hybrid rule-based and machine learning-based audit coding system developed for the neurosurgical department of a major trauma hospital.
    LanguageEnglish
    Title of host publicationProceedings of the 27th Australasian Conference on Information Systems (ACIS 2016)
    EditorsJulie Fisher, Walter Fernandez
    Place of PublicationWollongong, NSW, Australia
    PublisherUniversity of Wollongong
    Pages1-11
    Number of pages11
    ISBN (Electronic)9781741282672
    Publication statusPublished - 2016
    EventAustralasian Conference on Information Systems 2016: Occupying the Sweet Spot: IS at the Intersection - University of Wollongong, Wollongong, Australia
    Duration: 5 Dec 20167 Dec 2016
    Conference number: 27th
    http://business.uow.edu.au/acis-2016/index.html

    Conference

    ConferenceAustralasian Conference on Information Systems 2016
    Abbreviated titleACIS 2016
    CountryAustralia
    CityWollongong
    Period5/12/167/12/16
    OtherInformation systems (IS) have become an unrecognised commodity – everybody uses them, yet as IS researchers and practitioners we seem to need to explain time and again what we do, what value we provide, and keep justifying our existence.

    ACIS 2016 provides the opportunity to do just that and offers the opportunity how we, as the IS community, take up that challenge.
    Internet address

    Keywords

    • Neurosurgery
    • Information extraction
    • Rule-based expert systems
    • Machine learning
    • Audit coding

    Cite this

    Khademi , S., Delir Haghighi, P., Burstein, F., & Palmer, C. (2016). Enhancing rule-based text classification of neurosurgical notes using filtered feature weight vectors. In J. Fisher, & W. Fernandez (Eds.), Proceedings of the 27th Australasian Conference on Information Systems (ACIS 2016) (pp. 1-11). Wollongong, NSW, Australia: University of Wollongong.
    Khademi , Sedigheh ; Delir Haghighi, Pari ; Burstein, Frada ; Palmer, Christopher. / Enhancing rule-based text classification of neurosurgical notes using filtered feature weight vectors. Proceedings of the 27th Australasian Conference on Information Systems (ACIS 2016). editor / Julie Fisher ; Walter Fernandez. Wollongong, NSW, Australia : University of Wollongong, 2016. pp. 1-11
    @inproceedings{bbc3d701ee264d9e9fcc439a04c39ca8,
    title = "Enhancing rule-based text classification of neurosurgical notes using filtered feature weight vectors",
    abstract = "Clinicians need to record clinical encounters in written or spoken language, not only for its work-flow naturalness but also for its expressivity, precision, and capacity to convey all required information, which codified structured data is incapable of. Therefore, the structured data which is required for aggregation and analysis must be obtained from clinical text as a later step. Specialised areas of medicine use their own clinical language and clinical coding systems, resulting in unique challenges for the extraction process. Rule-based information extraction techniques have been used effectively in commercial systems and are favoured because they are easily understood and controlled. However, there is promising research into the use of machine learning techniques for extracting information,and this research explores the effectiveness of a hybrid rule-based and machine learning-based audit coding system developed for the neurosurgical department of a major trauma hospital.",
    keywords = "Neurosurgery, Information extraction, Rule-based expert systems, Machine learning, Audit coding",
    author = "Sedigheh Khademi and {Delir Haghighi}, Pari and Frada Burstein and Christopher Palmer",
    year = "2016",
    language = "English",
    pages = "1--11",
    editor = "Fisher, {Julie } and Walter Fernandez",
    booktitle = "Proceedings of the 27th Australasian Conference on Information Systems (ACIS 2016)",
    publisher = "University of Wollongong",

    }

    Khademi , S, Delir Haghighi, P, Burstein, F & Palmer, C 2016, Enhancing rule-based text classification of neurosurgical notes using filtered feature weight vectors. in J Fisher & W Fernandez (eds), Proceedings of the 27th Australasian Conference on Information Systems (ACIS 2016). University of Wollongong, Wollongong, NSW, Australia, pp. 1-11, Australasian Conference on Information Systems 2016, Wollongong, Australia, 5/12/16.

    Enhancing rule-based text classification of neurosurgical notes using filtered feature weight vectors. / Khademi , Sedigheh; Delir Haghighi, Pari; Burstein, Frada; Palmer, Christopher.

    Proceedings of the 27th Australasian Conference on Information Systems (ACIS 2016). ed. / Julie Fisher; Walter Fernandez. Wollongong, NSW, Australia : University of Wollongong, 2016. p. 1-11.

    Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

    TY - GEN

    T1 - Enhancing rule-based text classification of neurosurgical notes using filtered feature weight vectors

    AU - Khademi , Sedigheh

    AU - Delir Haghighi, Pari

    AU - Burstein, Frada

    AU - Palmer, Christopher

    PY - 2016

    Y1 - 2016

    N2 - Clinicians need to record clinical encounters in written or spoken language, not only for its work-flow naturalness but also for its expressivity, precision, and capacity to convey all required information, which codified structured data is incapable of. Therefore, the structured data which is required for aggregation and analysis must be obtained from clinical text as a later step. Specialised areas of medicine use their own clinical language and clinical coding systems, resulting in unique challenges for the extraction process. Rule-based information extraction techniques have been used effectively in commercial systems and are favoured because they are easily understood and controlled. However, there is promising research into the use of machine learning techniques for extracting information,and this research explores the effectiveness of a hybrid rule-based and machine learning-based audit coding system developed for the neurosurgical department of a major trauma hospital.

    AB - Clinicians need to record clinical encounters in written or spoken language, not only for its work-flow naturalness but also for its expressivity, precision, and capacity to convey all required information, which codified structured data is incapable of. Therefore, the structured data which is required for aggregation and analysis must be obtained from clinical text as a later step. Specialised areas of medicine use their own clinical language and clinical coding systems, resulting in unique challenges for the extraction process. Rule-based information extraction techniques have been used effectively in commercial systems and are favoured because they are easily understood and controlled. However, there is promising research into the use of machine learning techniques for extracting information,and this research explores the effectiveness of a hybrid rule-based and machine learning-based audit coding system developed for the neurosurgical department of a major trauma hospital.

    KW - Neurosurgery

    KW - Information extraction

    KW - Rule-based expert systems

    KW - Machine learning

    KW - Audit coding

    M3 - Conference Paper

    SP - 1

    EP - 11

    BT - Proceedings of the 27th Australasian Conference on Information Systems (ACIS 2016)

    A2 - Fisher, Julie

    A2 - Fernandez, Walter

    PB - University of Wollongong

    CY - Wollongong, NSW, Australia

    ER -

    Khademi S, Delir Haghighi P, Burstein F, Palmer C. Enhancing rule-based text classification of neurosurgical notes using filtered feature weight vectors. In Fisher J, Fernandez W, editors, Proceedings of the 27th Australasian Conference on Information Systems (ACIS 2016). Wollongong, NSW, Australia: University of Wollongong. 2016. p. 1-11