Alzheimer's Diagnosis Using Eigenbrains and Support Vector Machines
IWANN '09 Proceedings of the 10th International Work-Conference on Artificial Neural Networks: Part I: Bio-Inspired Systems: Computational and Ambient Intelligence
Pattern Recognition Letters
Computer aided diagnosis of Alzheimer's disease using component based SVM
Applied Soft Computing
| Hi-index | 0.00 |
Alzheimer's disease is a chronic degenerative diseaseof the central nervous system. Clinically early detection ofAlzheimer's disease is helpful in taking care of the patients.The nuclear imaging method, single-photon emission computedtomography (SPECT), is a useful tool in analyzingthe cerebral blood flow. Most common regional abnormalitiesfor Alzheimer's disease are symmetric or asymmetricbilateral temporal or parietal hypoperfusion, or frontal hypoperfusion.Statistical Parametric Mapping (SPM) is employedto do pre-processing of SPECT volumes. Due to itseffectiveness, easiness and fastness, SPM has been widelyapplied to the diagnosis and function research of brain diseases.The proposed system can provide a quantitatively automaticanalysis of the SPECT volumes. The selection of threevariables based on the statistical parametric t maps betweenAlzheimer's and normal volumes are proposed. Thenan optimal linear classifier is applied to discriminate betweenthese two group of volumes. In statistical patternrecognition, the Bayes error, the overlap among differentclass densities, is the smallest possible error in the currentmeasurement space. Due to the effectiveness of the variableselection, the simple optimal linear classifier achievesa near-Bayes error ratio. The sensitivity and specificity ofthe proposed method are 88% and 90%, respectively. Withthe high sensitivity and specificity performance, the proposedautomatic analysis of brain SPECT volumes can assistin the clinical practice of radiologists.