Gaussian Mixture Models and Model Selection for [18F] Fluorodeoxyglucose Positron Emission Tomography Classification in Alzheimer's Disease

Rui Li; Robert Perneczky; Igor Yakushev; Stefan Förster; Alexander Kurz; Alexander Drzezga; Stefan Kramer

doi:10.1371/journal.pone.0122731

Back

Gaussian Mixture Models and Model Selection for [18F] Fluorodeoxyglucose Positron Emission Tomography Classification in Alzheimer's Disease

Journal article

Open access

Peer reviewed

Gaussian Mixture Models and Model Selection for [18F] Fluorodeoxyglucose Positron Emission Tomography Classification in Alzheimer's Disease

Rui Li, Robert Perneczky, Igor Yakushev, Stefan Förster, Alexander Kurz, Alexander Drzezga and Stefan Kramer

PloS one, Vol.10(4), pp.e0122731-e0122731

2014

DOI: 10.1371/journal.pone.0122731

PMCID: PMC4412726

PMID: 25919662

Files and links (1)

url

https://doi.org/10.1371/journal.pone.0122731View

Published (Version of record) Open Access

Abstract

We present a method to discover discriminative brain metabolism patterns in [18F] fluorodeoxyglucose positron emission tomography (PET) scans, facilitating the clinical diagnosis of Alzheimer's disease. In the work, the term "pattern" stands for a certain brain region that characterizes a target group of patients and can be used for a classification as well as interpretation purposes. Thus, it can be understood as a so-called "region of interest (ROI)". In the literature, an ROI is often found by a given brain atlas that defines a number of brain regions, which corresponds to an anatomical approach. The present work introduces a semi-data-driven approach that is based on learning the characteristics of the given data, given some prior anatomical knowledge. A Gaussian Mixture Model (GMM) and model selection are combined to return a clustering of voxels that may serve for the definition of ROIs. Experiments on both an in-house dataset and data of the Alzheimer's Disease Neuroimaging Initiative (ADNI) suggest that the proposed approach arrives at a better diagnosis than a merely anatomical approach or conventional statistical hypothesis testing.

Models, Theoretical

Fluorodeoxyglucose F18 - pharmacokinetics

Brain - diagnostic imaging

Radiopharmaceuticals - pharmacokinetics

Humans

Male

Positron-Emission Tomography - methods

Normal Distribution

Alzheimer Disease - pathology

Sensitivity and Specificity

Aged, 80 and over

Brain - pathology

Female

Aged

Alzheimer Disease - diagnostic imaging

Details

Title: Subtitle: Gaussian Mixture Models and Model Selection for [18F] Fluorodeoxyglucose Positron Emission Tomography Classification in Alzheimer's Disease
Creators: Rui Li - Institut für Informatik/I12, Technische Universität München, Garching bei München, Germany
Robert Perneczky - Klinik und Poliklinik für Psychiatrie und Psychotherapie, Technische Universität München, München, Germany
Igor Yakushev - Nuklearmedizinische Klinik, Technische Universität München, München, Germany
Stefan Förster - Nuklearmedizinische Klinik, Technische Universität München, München, Germany
Alexander Kurz - Klinik und Poliklinik für Psychiatrie und Psychotherapie, Technische Universität München, München, Germany
Alexander Drzezga - Klinik und Poliklinik für Nuklearmedizin, Universität zu Köln, Köln, Germany
Stefan Kramer - Institut für Informatik, Johannes Gutenberg-Universität Mainz, Mainz, Germany
Resource Type: Journal article
Publication Details: PloS one, Vol.10(4), pp.e0122731-e0122731
DOI: 10.1371/journal.pone.0122731
PMID: 25919662
PMCID: PMC4412726
NLM abbreviation: PLoS One
ISSN: 1932-6203
eISSN: 1932-6203
Publisher: Public Library of Science; United States
Grant note: P30 AG062421 / NIA NIH HHS P50 AG005134 / NIA NIH HHS MR/M024903/1 / Medical Research Council U01 AG024904 / NIA NIH HHS
Language: English
Date published: 2014
Academic Unit: Neurology
Record Identifier: 9984051758602771

Metrics

13 Record Views

23 Times Cited - Web of Science