Title :
3D brain image-based diagnosis of Alzheimer´s disease: Bringing medical vision into feature selection
Author :
Bicacro, Eduardo ; Silveira, Margarida ; Marques, Jorge S. ; Costa, Durval C.
Author_Institution :
Inst. Super. Tecnico, Lisbon, Portugal
Abstract :
Expert physicians are able to attain good Alzheimer´s Disease (AD) diagnostic accuracy, relying on visual inspection of Positron Emission Tomography (PET) images only. Nevertheless, computerized methods have been implemented with similar or even better performance. We investigate the potential of the physician´s experienced visual inspection to guide feature selection, in an automatic classification procedure. Eye tracking methodology is employed to obtain a model of the physician´s visual behavior, which allows for the sampling of voxel intensity features that are then fed to an SVM classifier. This approach is compared with commonly used automatic feature selection alternatives. Image data were taken from the Alzheimer´s Disease Neuroimaging Initiative database. The results show that the proposed approach marginally improves accuracy in AD vs. CN classification, but for MCI vs. CN and AD vs. MCI it presents lower performance.
Keywords :
brain; diseases; feature extraction; image classification; medical image processing; positron emission tomography; support vector machines; vision; 3D brain image-based diagnosis; Alzheimer disease neuroimaging initiative database; PET; SVM classifier; automatic classification procedure; diagnostic accuracy; eye tracking methodology; feature selection; medical vision; positron emission tomography; support vector machines; voxel intensity features; Alzheimer´s disease; Biomedical imaging; Positron emission tomography; Support vector machines; Visualization; Alzheimer´s disease; Computer aided diagnosis; Eye tracking; Feature selection; Positron Emission Tomography;
Conference_Titel :
Biomedical Imaging (ISBI), 2012 9th IEEE International Symposium on
Conference_Location :
Barcelona
Print_ISBN :
978-1-4577-1857-1
DOI :
10.1109/ISBI.2012.6235502