Syntactic Similarity Of Medical Reports

the similarity of medical reports is evaluated by calculating the semantic characteristics and syntactic similarity. It relies on an upgraded radiology-specific ontology to measure semantic similarity relationships between unstructured mammographic report concepts. While [7] improved the vector cosine similarity algorithm model which uses (is-a) relationships to measure the degree of similarity. For a fixed concept, after examining all the possible paths they arrived at the conclusion that the shortest similarity vector would be selected for each document then the cosine angle of each vector is calculated to determine the degree of similarity. testing has been done by comparing multiple clinical context reports using anatomy and imaging procedures