International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 69 - Number 5 |
Year of Publication: 2013 |
Authors: Valanarasi Antony Santiagu Vaz |
10.5120/11839-7567 |
Valanarasi Antony Santiagu Vaz . Early Diagnosis of Parkinson’s Disease based on Transcranial Sonography. International Journal of Computer Applications. 69, 5 ( May 2013), 22-28. DOI=10.5120/11839-7567
Transcranial Sonography (TCS) plays a very significant role in the early analysis of Parkinson's disease (PD). The TCS taken in the mesencephelon region shows a discrete pattern with increase in size (hyper echogenecity) of substantia nigra in about 90% of PD patients. Generally this hyperechogenic pattern is segmented physically which can be used as PD indicator for early diagnosis. This paper proposes a novel procedure using GLCM and Multi Layer Perceptron Neural Network for the early PD risk assessment. The features are obtained by a assortment of Gabor filters, and the concert of these features is evaluated by feature selection method. At an earlier stage speckle noise is removed using spatially adaptive wiener filter. This method is well applicable with neural network toolbox in MATLAB.