Classifying Volume Datasets Based on Intensities and Geometric Features

Abstract Many state-of-the art visualization techniques must be tailored to the spe-cific type of dataset, its modality (CT, MRI, etc.), the recorded object or anatomical region (head, spine, abdomen, etc.) and other parameters related to the data acqui-sition process. While parts of the information (imaging modality and acquisition sequence) may be obtained from the meta-data stored with the volume scan, there is important information which is not stored explicitly, e.g. anatomical region. Also, meta-data might be incomplete, inappropriate or simply missing. This paper presents a novel and simple method of determining the type of dataset from previously defined categories. A 2D histogram of the dataset is used as input to the neural network, which classifies it into one of several categories it was trained with. Two types of 2D histograms have been experimented with, one based on in-tensity and gradient magnitude, the other one on intensity and distance from center. A significant result is the ability of the system to classify datasets into a specific class after being trained with only one dataset of that class. Other advantages of the method are its easy implementation and its high computational performance