Feature selection and time regression software: Application on predicting Alzheimer's disease progress

Dimitrios Ververidis, Mark van Gils, Juha Koikkalainen, Jyrki Lötjönen

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

    3 Citations (Scopus)

    Abstract

    In this paper, the Bayes classifier is used to predict Alzheimer's disease progress. The classifier is trained on a subset of the Alzheimer's Disease Neuroimaging Initiative database. Subjects are diagnosed by doctors as belonging to healthy, mild-cognitive impaired, and Alzheimer's disease class. A software tool for features selection and time regression is developed. The tool utilizes a variant of the Sequential Forward Selection (SFS) algorithm for feature selection, where the criterion used for selecting features is the correct classification rate of the Bayes classifier. The tool also employs linear regression to predict future values of selected biomarkers, such as the hippocampus volume, from past measurements, so that future class of the subject can be predicted.

    Original languageEnglish
    Title of host publicationProceedings of EUSIPCO 2010
    PublisherEuropean Association for Signal and Image Processing (EURASIP)
    Pages1179-1183
    Publication statusPublished - 1 Dec 2010
    MoE publication typeNot Eligible
    Event18th European Signal Processing Conference, EUSIPCO 2010 - Aalborg, Denmark
    Duration: 23 Aug 201027 Aug 2010

    Conference

    Conference18th European Signal Processing Conference, EUSIPCO 2010
    CountryDenmark
    CityAalborg
    Period23/08/1027/08/10

    Keywords

    • feature selection
    • Alzheimer`s disease
    • pattern recognition
    • decision support
    • data mining

    Fingerprint Dive into the research topics of 'Feature selection and time regression software: Application on predicting Alzheimer's disease progress'. Together they form a unique fingerprint.

    Cite this