Computer Aided Diagnosis for mental health care: On the Clinical Validation of Sensitive Machines

Frans van der Sluis, Ton Dijkstra, Egon van den Broek

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

    2 Citations (Scopus)
    23 Downloads (Pure)

    Abstract

    This study explores the feasibility of sensitive machines; that is, machines with empathic abilities, at least to some extent. A signal processing and machine learning pipeline is presented that is used to analyze data from two studies in which 25 Post-Traumatic Stress Disorder (PTSD) patients participated. The feasibility of speech as a stress detector was validated in a clinical setting, using the Subjective Unit of Distress (SUD). 13 statistical parameters were derived from five speech features, namely: amplitude, zero crossings, power, high-frequency power, and pitch. To achieve a low dimensional representation, a subset of 28 parameters was selected and, subsequently, compressed into 11 principal components (PC). Using a Multi-Layer Perceptron neural network (MLP), the set of 11 PC were mapped upon 9 distinct quantizations of the SUD. The MLP was able to discriminate between 2 stress levels with 82.4% accuracy and up to 10 stress levels with 36.3% accuracy. With stress baptized as being the black death of the 21st century, this work can be conceived as an important step towards computer aided mental health care.
    Original languageUndefined
    Title of host publicationProceedings of the International Conference on Health Informatics, HealthInf 2012
    EditorsE. Conchon, C. Correia, A. Fred, H. Gamboa
    Place of PublicationPortugal
    PublisherSCITEPRESS - Science and Technology Publications
    Pages493-498
    Number of pages7
    ISBN (Print)978-989-8425-88-1
    Publication statusPublished - 1 Feb 2012

    Publication series

    Name
    PublisherSciTePress - Science and Technology Publications

    Keywords

    • METIS-285126
    • IR-79668
    • Stress
    • Speech
    • Validation
    • HMI-HF: Human Factors
    • Mental healthcare
    • Computer Aided Diagnostics (CAD)
    • EWI-21504
    • HMI-SLT: Speech and Language Technology
    • artificial neural network

    Cite this

    van der Sluis, F., Dijkstra, T., & van den Broek, E. (2012). Computer Aided Diagnosis for mental health care: On the Clinical Validation of Sensitive Machines. In E. Conchon, C. Correia, A. Fred, & H. Gamboa (Eds.), Proceedings of the International Conference on Health Informatics, HealthInf 2012 (pp. 493-498). Portugal: SCITEPRESS - Science and Technology Publications.
    van der Sluis, Frans ; Dijkstra, Ton ; van den Broek, Egon. / Computer Aided Diagnosis for mental health care: On the Clinical Validation of Sensitive Machines. Proceedings of the International Conference on Health Informatics, HealthInf 2012. editor / E. Conchon ; C. Correia ; A. Fred ; H. Gamboa. Portugal : SCITEPRESS - Science and Technology Publications, 2012. pp. 493-498
    @inproceedings{9b49f1b921724dfe94440ee6b55ed017,
    title = "Computer Aided Diagnosis for mental health care: On the Clinical Validation of Sensitive Machines",
    abstract = "This study explores the feasibility of sensitive machines; that is, machines with empathic abilities, at least to some extent. A signal processing and machine learning pipeline is presented that is used to analyze data from two studies in which 25 Post-Traumatic Stress Disorder (PTSD) patients participated. The feasibility of speech as a stress detector was validated in a clinical setting, using the Subjective Unit of Distress (SUD). 13 statistical parameters were derived from five speech features, namely: amplitude, zero crossings, power, high-frequency power, and pitch. To achieve a low dimensional representation, a subset of 28 parameters was selected and, subsequently, compressed into 11 principal components (PC). Using a Multi-Layer Perceptron neural network (MLP), the set of 11 PC were mapped upon 9 distinct quantizations of the SUD. The MLP was able to discriminate between 2 stress levels with 82.4{\%} accuracy and up to 10 stress levels with 36.3{\%} accuracy. With stress baptized as being the black death of the 21st century, this work can be conceived as an important step towards computer aided mental health care.",
    keywords = "METIS-285126, IR-79668, Stress, Speech, Validation, HMI-HF: Human Factors, Mental healthcare, Computer Aided Diagnostics (CAD), EWI-21504, HMI-SLT: Speech and Language Technology, artificial neural network",
    author = "{van der Sluis}, Frans and Ton Dijkstra and {van den Broek}, Egon",
    year = "2012",
    month = "2",
    day = "1",
    language = "Undefined",
    isbn = "978-989-8425-88-1",
    publisher = "SCITEPRESS - Science and Technology Publications",
    pages = "493--498",
    editor = "E. Conchon and C. Correia and A. Fred and H. Gamboa",
    booktitle = "Proceedings of the International Conference on Health Informatics, HealthInf 2012",

    }

    van der Sluis, F, Dijkstra, T & van den Broek, E 2012, Computer Aided Diagnosis for mental health care: On the Clinical Validation of Sensitive Machines. in E Conchon, C Correia, A Fred & H Gamboa (eds), Proceedings of the International Conference on Health Informatics, HealthInf 2012. SCITEPRESS - Science and Technology Publications, Portugal, pp. 493-498.

    Computer Aided Diagnosis for mental health care: On the Clinical Validation of Sensitive Machines. / van der Sluis, Frans; Dijkstra, Ton; van den Broek, Egon.

    Proceedings of the International Conference on Health Informatics, HealthInf 2012. ed. / E. Conchon; C. Correia; A. Fred; H. Gamboa. Portugal : SCITEPRESS - Science and Technology Publications, 2012. p. 493-498.

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

    TY - GEN

    T1 - Computer Aided Diagnosis for mental health care: On the Clinical Validation of Sensitive Machines

    AU - van der Sluis, Frans

    AU - Dijkstra, Ton

    AU - van den Broek, Egon

    PY - 2012/2/1

    Y1 - 2012/2/1

    N2 - This study explores the feasibility of sensitive machines; that is, machines with empathic abilities, at least to some extent. A signal processing and machine learning pipeline is presented that is used to analyze data from two studies in which 25 Post-Traumatic Stress Disorder (PTSD) patients participated. The feasibility of speech as a stress detector was validated in a clinical setting, using the Subjective Unit of Distress (SUD). 13 statistical parameters were derived from five speech features, namely: amplitude, zero crossings, power, high-frequency power, and pitch. To achieve a low dimensional representation, a subset of 28 parameters was selected and, subsequently, compressed into 11 principal components (PC). Using a Multi-Layer Perceptron neural network (MLP), the set of 11 PC were mapped upon 9 distinct quantizations of the SUD. The MLP was able to discriminate between 2 stress levels with 82.4% accuracy and up to 10 stress levels with 36.3% accuracy. With stress baptized as being the black death of the 21st century, this work can be conceived as an important step towards computer aided mental health care.

    AB - This study explores the feasibility of sensitive machines; that is, machines with empathic abilities, at least to some extent. A signal processing and machine learning pipeline is presented that is used to analyze data from two studies in which 25 Post-Traumatic Stress Disorder (PTSD) patients participated. The feasibility of speech as a stress detector was validated in a clinical setting, using the Subjective Unit of Distress (SUD). 13 statistical parameters were derived from five speech features, namely: amplitude, zero crossings, power, high-frequency power, and pitch. To achieve a low dimensional representation, a subset of 28 parameters was selected and, subsequently, compressed into 11 principal components (PC). Using a Multi-Layer Perceptron neural network (MLP), the set of 11 PC were mapped upon 9 distinct quantizations of the SUD. The MLP was able to discriminate between 2 stress levels with 82.4% accuracy and up to 10 stress levels with 36.3% accuracy. With stress baptized as being the black death of the 21st century, this work can be conceived as an important step towards computer aided mental health care.

    KW - METIS-285126

    KW - IR-79668

    KW - Stress

    KW - Speech

    KW - Validation

    KW - HMI-HF: Human Factors

    KW - Mental healthcare

    KW - Computer Aided Diagnostics (CAD)

    KW - EWI-21504

    KW - HMI-SLT: Speech and Language Technology

    KW - artificial neural network

    M3 - Conference contribution

    SN - 978-989-8425-88-1

    SP - 493

    EP - 498

    BT - Proceedings of the International Conference on Health Informatics, HealthInf 2012

    A2 - Conchon, E.

    A2 - Correia, C.

    A2 - Fred, A.

    A2 - Gamboa, H.

    PB - SCITEPRESS - Science and Technology Publications

    CY - Portugal

    ER -

    van der Sluis F, Dijkstra T, van den Broek E. Computer Aided Diagnosis for mental health care: On the Clinical Validation of Sensitive Machines. In Conchon E, Correia C, Fred A, Gamboa H, editors, Proceedings of the International Conference on Health Informatics, HealthInf 2012. Portugal: SCITEPRESS - Science and Technology Publications. 2012. p. 493-498