18F-FDG PET brain images as features for Alzheimer classification. (August 2017)

• Record Type: Journal Article
• Authors: Azmi, M.H.; Saripan, M.I.; Nordin, A.J.; Ahmad Saad, F.F.; Abdul Aziz, S.A.; Wan Adnan, W.A.
• Is Part Of: Radiation physics and chemistry. Volume 137(2017)
• Page Start: 135
• Page End: 143
• Publication Date: 2017-08
• Subjects: PET -- Alzheimer -- Automatic classification -- Z-score -- Image processing; Radiation chemistry -- Periodicals; Radiometry -- Periodicals; Radiation -- Periodicals; Chimie sous rayonnement -- Périodiques; 539.2
• DOI: 10.1016/j.radphyschem.2016.08.028
• Languages: English
• ISSNs: 0969-806X

Abstract

Abstract: 2-Deoxy-2-[fluorine-18] fluoro-D -glucose ( 18 F-FDG) Positron Emission Tomography (PET) imaging offers meaningful information for various types of diseases diagnosis. In Alzheimer's disease (AD), the hypometabolism of glucose which observed on the low intensity voxel in PET image may relate to the onset of the disease. The importance of early detection of AD is inevitable because the resultant brain damage is irreversible. Several statistical analysis and machine learning algorithm have been proposed to investigate the rate and the pattern of the hypometabolism. This study focus on the same aim with further investigation was performed on several hypometabolism pattern. Some pre-processing steps were implemented to standardize the data in order to minimize the effect of resolution and anatomical differences. The features used are the mean voxel intensity within the AD pattern mask, which derived from several z-score and FDR threshold values. The global mean voxel (GMV) and slice-based mean voxel (SbMV) intensity were observed and used as input to the neural network. Several neural network architectures were tested and compared to the nearest neighbour method. The highest accuracy equals to 0.9 and recorded at z-score ≤−1.3 with 1 node neural network architecture (sensitivity=0.81 and specificity=0.95) and at z-score ≤−0.7 with 10 nodes neural network (sensitivity=0.83 and specificity=0.94). Highlights: Investigate global and slice-based z-score and FDR as features to classify Alzheimer disease. Investigate optimal threshold of z-score and FDR to select significant voxels to discriminate Alzheimer disease. Investigate the potential of neural network as classifier in Alzheimer disease classification.