Medical Image classification based on feature extraction

Hi,

I have done segmentation of Medical image(Brain Image). I have taken around 90 images(55 normal and 35 Abnormal) and have extracted 20 features for each image (20*90=1800 features). I have kept the features in a single Excel file.Now i need to classify the image as normal or abnormal. Can i go with Svm or Knn?What is the procedure i have to follow in this and is there any inbuilt codes for this in Matlab. I am using 2014 a regulation of Matlab.Please help...

[KALYAN ACHARJYA]

Yes, how you decided the features?

Is there any significant difference on features to classify, whether the image is normal or abnormal?

[Ramanathan Anandarama Krishnan]

Hi,

I extracted features like area, mean,S.D.,entropy.glcm etc and there is a close relation between normal and abnormal images.In some images area is more in abnormal images(Actually it should be less). However all values are not same with all the other values and there is a slight difference in normal and abnormal values...