Automated bone density calculation using feature extraction by deformable templates

The authors propose a method for detecting and describing features in medical images by using deformable templates for the purpose of diagnostic analysis of these features. The feature of interest can be described by a parameterized template. A gradient maximization technique is used to change the parameters so that the template fits onto the desired anatomical feature. The final parameter values can be used as descriptors for the feature. Measurements of intensity values within a region of the template can be used as inputs to a medical diagnostic system. A picture-archive-and-communication-system-based image analysis program, which uses the technique of deformable templates to localize features in dual energy computer tomography (CT) images, is discussed. Statistical comparisons between the system and previously used manual techniques show that the performances are practically equivalent and that the system has several advantages over the human operator, for example, consistency, accuracy, and cost