Classification of single-trial MEG during sentence processing for automated schizophrenia screening

Tingting Xu, Massoud Stephane, Keshab K. Parhi

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    3 Citations (Scopus)

    Abstract

    This paper presents a novel computer-aided system for assisting schizophrenia (SZ) diagnosis. Power Spectral Density Ratios (PSDRs) covering 7 brain regions and 5 frequency sub-bands are extracted as features, from single-trial magnetoencephalography (MEG) recorded while subjects read sentence stimuli silently. A two-stage feature selection algorithm combining F-score and Adaptive Boosting (Adaboost) model is proposed to rank the features. The top ranked features are used to build a boosted non-linear classifier using linear decision stumps as the base classifiers. A majority voting scheme is employed to combine single trial classification results from each test subject to make final classification decisions. Following a leave-one-out cross validation procedure, the proposed system achieves 82.61% classification accuracy (92.31% specificity and 70% sensitivity) on 13 healthy controls and 10 SZ patients. The most discriminating PSDR features are selected from the right temporal, right parietal and right frontal regions and are related to alpha (8-13Hz) and beta (13-30Hz) frequency ranges. This information may help in gaining knowledge about the abnormal neural oscillations associated with sentence-level language disorder in SZ.

    Original languageEnglish (US)
    Title of host publicationInternational IEEE/EMBS Conference on Neural Engineering, NER
    Pages363-366
    Number of pages4
    DOIs
    StatePublished - 2013
    Event2013 6th International IEEE EMBS Conference on Neural Engineering, NER 2013 - San Diego, CA, United States
    Duration: Nov 6 2013Nov 8 2013

    Other

    Other2013 6th International IEEE EMBS Conference on Neural Engineering, NER 2013
    CountryUnited States
    CitySan Diego, CA
    Period11/6/1311/8/13

    Fingerprint

    Magnetoencephalography
    Screening
    Power spectral density
    Classifiers
    Processing
    Adaptive boosting
    Feature extraction
    Brain

    Keywords

    • Adaboost
    • Classification
    • Computer-aided schizophrenia identification
    • Feature selection
    • Magnetoencephalography (MEG)
    • Power spectral density ratio (PSDR)

    ASJC Scopus subject areas

    • Artificial Intelligence
    • Mechanical Engineering

    Cite this

    Xu, T., Stephane, M., & Parhi, K. K. (2013). Classification of single-trial MEG during sentence processing for automated schizophrenia screening. In International IEEE/EMBS Conference on Neural Engineering, NER (pp. 363-366). [6695947] https://doi.org/10.1109/NER.2013.6695947

    Classification of single-trial MEG during sentence processing for automated schizophrenia screening. / Xu, Tingting; Stephane, Massoud; Parhi, Keshab K.

    International IEEE/EMBS Conference on Neural Engineering, NER. 2013. p. 363-366 6695947.

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Xu, T, Stephane, M & Parhi, KK 2013, Classification of single-trial MEG during sentence processing for automated schizophrenia screening. in International IEEE/EMBS Conference on Neural Engineering, NER., 6695947, pp. 363-366, 2013 6th International IEEE EMBS Conference on Neural Engineering, NER 2013, San Diego, CA, United States, 11/6/13. https://doi.org/10.1109/NER.2013.6695947
    Xu T, Stephane M, Parhi KK. Classification of single-trial MEG during sentence processing for automated schizophrenia screening. In International IEEE/EMBS Conference on Neural Engineering, NER. 2013. p. 363-366. 6695947 https://doi.org/10.1109/NER.2013.6695947
    Xu, Tingting ; Stephane, Massoud ; Parhi, Keshab K. / Classification of single-trial MEG during sentence processing for automated schizophrenia screening. International IEEE/EMBS Conference on Neural Engineering, NER. 2013. pp. 363-366
    @inproceedings{9bc34c0520cf48539ae7d14512979474,
    title = "Classification of single-trial MEG during sentence processing for automated schizophrenia screening",
    abstract = "This paper presents a novel computer-aided system for assisting schizophrenia (SZ) diagnosis. Power Spectral Density Ratios (PSDRs) covering 7 brain regions and 5 frequency sub-bands are extracted as features, from single-trial magnetoencephalography (MEG) recorded while subjects read sentence stimuli silently. A two-stage feature selection algorithm combining F-score and Adaptive Boosting (Adaboost) model is proposed to rank the features. The top ranked features are used to build a boosted non-linear classifier using linear decision stumps as the base classifiers. A majority voting scheme is employed to combine single trial classification results from each test subject to make final classification decisions. Following a leave-one-out cross validation procedure, the proposed system achieves 82.61{\%} classification accuracy (92.31{\%} specificity and 70{\%} sensitivity) on 13 healthy controls and 10 SZ patients. The most discriminating PSDR features are selected from the right temporal, right parietal and right frontal regions and are related to alpha (8-13Hz) and beta (13-30Hz) frequency ranges. This information may help in gaining knowledge about the abnormal neural oscillations associated with sentence-level language disorder in SZ.",
    keywords = "Adaboost, Classification, Computer-aided schizophrenia identification, Feature selection, Magnetoencephalography (MEG), Power spectral density ratio (PSDR)",
    author = "Tingting Xu and Massoud Stephane and Parhi, {Keshab K.}",
    year = "2013",
    doi = "10.1109/NER.2013.6695947",
    language = "English (US)",
    isbn = "9781467319690",
    pages = "363--366",
    booktitle = "International IEEE/EMBS Conference on Neural Engineering, NER",

    }

    TY - GEN

    T1 - Classification of single-trial MEG during sentence processing for automated schizophrenia screening

    AU - Xu, Tingting

    AU - Stephane, Massoud

    AU - Parhi, Keshab K.

    PY - 2013

    Y1 - 2013

    N2 - This paper presents a novel computer-aided system for assisting schizophrenia (SZ) diagnosis. Power Spectral Density Ratios (PSDRs) covering 7 brain regions and 5 frequency sub-bands are extracted as features, from single-trial magnetoencephalography (MEG) recorded while subjects read sentence stimuli silently. A two-stage feature selection algorithm combining F-score and Adaptive Boosting (Adaboost) model is proposed to rank the features. The top ranked features are used to build a boosted non-linear classifier using linear decision stumps as the base classifiers. A majority voting scheme is employed to combine single trial classification results from each test subject to make final classification decisions. Following a leave-one-out cross validation procedure, the proposed system achieves 82.61% classification accuracy (92.31% specificity and 70% sensitivity) on 13 healthy controls and 10 SZ patients. The most discriminating PSDR features are selected from the right temporal, right parietal and right frontal regions and are related to alpha (8-13Hz) and beta (13-30Hz) frequency ranges. This information may help in gaining knowledge about the abnormal neural oscillations associated with sentence-level language disorder in SZ.

    AB - This paper presents a novel computer-aided system for assisting schizophrenia (SZ) diagnosis. Power Spectral Density Ratios (PSDRs) covering 7 brain regions and 5 frequency sub-bands are extracted as features, from single-trial magnetoencephalography (MEG) recorded while subjects read sentence stimuli silently. A two-stage feature selection algorithm combining F-score and Adaptive Boosting (Adaboost) model is proposed to rank the features. The top ranked features are used to build a boosted non-linear classifier using linear decision stumps as the base classifiers. A majority voting scheme is employed to combine single trial classification results from each test subject to make final classification decisions. Following a leave-one-out cross validation procedure, the proposed system achieves 82.61% classification accuracy (92.31% specificity and 70% sensitivity) on 13 healthy controls and 10 SZ patients. The most discriminating PSDR features are selected from the right temporal, right parietal and right frontal regions and are related to alpha (8-13Hz) and beta (13-30Hz) frequency ranges. This information may help in gaining knowledge about the abnormal neural oscillations associated with sentence-level language disorder in SZ.

    KW - Adaboost

    KW - Classification

    KW - Computer-aided schizophrenia identification

    KW - Feature selection

    KW - Magnetoencephalography (MEG)

    KW - Power spectral density ratio (PSDR)

    UR - http://www.scopus.com/inward/record.url?scp=84897689426&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84897689426&partnerID=8YFLogxK

    U2 - 10.1109/NER.2013.6695947

    DO - 10.1109/NER.2013.6695947

    M3 - Conference contribution

    SN - 9781467319690

    SP - 363

    EP - 366

    BT - International IEEE/EMBS Conference on Neural Engineering, NER

    ER -