Automatic Detection of ADHD based on the Selected Features from PET Scan Images

Developing an exact and efficient automated system for the initial detection of Alzheimer’s Disease (AD) is of main importance for effective treatment. Recently, there has been great interest in Computer Aided Diagnosis (CAD) system for AD. However, distinguishing normal control from Aided Diagnosis patients is a very difficult work due to their intensities and patterns. There are many applications are found out in statistical research in many fields including medical. In this paper, we develop a fully automatic CAD system for the diagnosis of AD from PET images. Image features are derived by attaining Discrete Wavelet Transform (DWT). Subsequently, the way of look is decreased by computing some statistical feature vectors. Then the decreased feature set is passed to Multilayer Perceptron (MLP) to distinguish normal control and AD from PET image. Newly invented method is utilized, examined and compared with other old method in terms of classification accuracy, precise and carefulness. Results exactly say that the proposed CAD system is worth than other CAD methods reported in the literature. Additionally, the proposed CAD can be used as a diagnostic tool for AD with the ability of defining initial stages of the disease.