Application of the predictad software tool to predict progression in patients with mild cognitive impairment

Anja H. Simonsen (Corresponding Author), Jussi Mattila, Anne Mette Hejl, Kristian S. Frederiksen, Sanna Kaisa Herukka, Merja Hallikainen, Mark van Gils, Jyrki Lötjönen, Hilkka Soininen, Gunhild Waldemar

    Research output: Contribution to journalArticleScientificpeer-review

    11 Citations (Scopus)


    Background: The PredictAD tool integrates heterogeneous data such as imaging, cerebrospinal fluid biomarkers and results from neuropsychological tests for compact visualization in an interactive user interface. This study investigated whether the software tool could assist physicians in the early diagnosis of Alzheimer's disease. Methods: Baseline data from 140 patients with mild cognitive impairment were selected from the Alzheimer's Disease Neuroimaging Study. Three clinical raters classified patients into 6 categories of confidence in the prediction of early Alzheimer's disease, in 4 phases of incremental data presentation using the software tool. A 5th phase was done with all available patient data presented on paper charts. Classifications by the clinical raters were compared to the clinical diagnoses made by the Alzheimer's Disease Neuroimaging Initiative investigators. Results: A statistical significant trend (p < 0.05) towards better classification accuracy (from 62.6 to 70.0%) was found when using the PredictAD tool during the stepwise procedure. When the same data were presented on paper, classification accuracy of the raters dropped significantly from 70.0 to 63.2%. Conclusion: Best classification accuracy was achieved by the clinical raters when using the tool for decision support, suggesting that the tool can add value in diagnostic classification when large amounts of heterogeneous data are presented.

    Original languageEnglish
    Pages (from-to)344-350
    JournalDementia and Geriatric Cognitive Disorders
    Issue number5-6
    Publication statusPublished - 2012
    MoE publication typeA1 Journal article-refereed


    • Alzheimer's disease
    • Mild cognitive impairment
    • PredictAD software tool


    Dive into the research topics of 'Application of the predictad software tool to predict progression in patients with mild cognitive impairment'. Together they form a unique fingerprint.

    Cite this