Skip to main content
. Author manuscript; available in PMC: 2018 Mar 28.
Published in final edited form as: Proc SPIE Int Soc Opt Eng. 2016 Mar;9786:978605. doi: 10.1117/12.2216315

Algorithm, part 3.

Classification rule for new data.

input test sub-images, dictionary, model.
output label (calcium, non-calcium)
For each new sub-image {

  1. Convolve with filter bank, F and record convolution value.

  2. Combine with DGAS feature set (ℝn for each pixel)

  3. Label each pixel response by comparing to the dictionary (using k-NN)

  4. Create a histogram of label frequencies

  5. Normalize to get distribution (hence, sub-images do not need to be the same size)

  6. Predict using OC-SVM model

}