A Machine Learning Approach for Specification of Spinal Cord Injuries Using Fractional Anisotropy Values Obtained from Diffusion Tensor Images

Algorithm 1

Pseudocode for KNN classification.

𝑘-Nearest Neighbor

Classify (𝐗,𝐘,𝑥) // 𝐗: training data, 𝐘: class labels of 𝐗, 𝑥: unknown sample

for  𝑖=1  to  𝑚  do

 Compute distance 𝑑(𝐗_𝑖,𝑥)

end for

Compute set I containing indices for the k smallest distances 𝑑(𝐗_𝑖,𝑥).

return majority label for {𝐘_𝑖 where 𝑖∈𝐼}