• DocumentCode
    3532374
  • Title

    Machine learning for very early Alzheimer´s Disease diagnosis; a 18F-FDG and PiB PET comparison

  • Author

    Illan, I.A. ; Górriz, J.M. ; Ramírez, J. ; Chaves, R. ; Segovia, F. ; López, M. ; Salas-Gonzalez, D. ; Padilla, P. ; Puntonet, C.G.

  • Author_Institution
    Dept. of Signal Theor., Networking & Commun., Univ. of Granada, Granada, Spain
  • fYear
    2010
  • fDate
    Oct. 30 2010-Nov. 6 2010
  • Firstpage
    2334
  • Lastpage
    2337
  • Abstract
    This paper shows a machine learning approach based on Principal Component Analysis (PCA) and Support Vector Machine (SVM) to compare the diagnostic accuracy on very early Alzheimer´s Disease (AD) patients with 18F FDG and Pittsburg Compound B (PiB) PET imaging. The Alzheimer´s Disease Neuroimaging Initiative (ADNI) dataset is used for testing, making use of the longitudinal character. Mild Cognitive Impairment (MCI) individuals that after a two years follow up converted into possible AD where used as very early AD patients. While 18F FDG and PiB have similar diagnostic accuracy in AD, PiB is shown to have higher discriminative power in very early AD with respect to FDG.
  • Keywords
    diseases; learning (artificial intelligence); medical image processing; neurophysiology; positron emission tomography; principal component analysis; support vector machines; 18F-FDG PET; Alzheimer Disease Neuroimaging Initiative dataset; PCA; PiB PET; Pittsburg Compound B PET imaging; SVM; machine learning; mild cognitive impairment; principal component analysis; support vector machine; very early Alzheimer disease diagnosis; Accuracy; Alzheimer´s disease; Kernel; Positron emission tomography; Principal component analysis; Sensitivity; Support vector machines;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Nuclear Science Symposium Conference Record (NSS/MIC), 2010 IEEE
  • Conference_Location
    Knoxville, TN
  • ISSN
    1095-7863
  • Print_ISBN
    978-1-4244-9106-3
  • Type

    conf

  • DOI
    10.1109/NSSMIC.2010.5874201
  • Filename
    5874201