Application of decision tree for MRI images of premature brain injury classification

摘要

An optimization classification algorithm for MRI images of premature brain injury is introduced. Based on the shortcomings of the classical ID3 algorithm in dealing with the continuous attributes of medical image, the new algorithm selects the testing feature by comparing the information gain ratio and adds the handling methods for filling null values. Then it discrete the continuous attributes by dividing them into segments to classify the object. The result shows that the new algorithm can accurately classify the MRI images.