Aboul Ella Hassanien (Abo)

Extraction of relevant features is of significant importance for brain tumor segmentation systems. To improve brain tumor segmentation accuracy, the authors present an improved feature extraction component that takes advantage of the correlation between intracranial structure deformation and the compression resulting from brain tumor growth. Using 3D nonrigid registration and deformation modeling techniques, the component measures lateral ventricular (LaV) deformation in volumetric magnetic resonance images. By verifying the location of the extracted LaV deformation feature data and applying the features on brain tumor segmentation with widely used classification algorithms, the authors evaluate the proposed component qualitatively and quantitatively with promising results on 11 datasets comprising real and simulated patient images.

Abstract—Extraction of relevant features is of significant importance for brain tumor segmentation systems. In this paper, with the objective of improving brain tumor segmentation accuracy, we present an improved feature extraction component to take advantage of the correlation between intracranial structure deformation and the compression from brain tumor growth. Using 3-dimensional non-rigid registration and deformation modeling techniques, the component is capable of measuring lateral ventricular (LaV) deformation in the volumetric magnetic resonance (MR) images. By verifying the location of the extracted LaV deformation feature data and applying the features on brain tumor segmentation with widely used classification algorithms, the proposed component is evaluated qualitatively and quantitatively with promising results on 11 datasets comprising real patient and simulated images.

The Taylor expansion has shown in many fields to be an extremely powerful tool. In this paper, we investigated image features and their relationships by analogy with Taylor expansion. The kind of expansion could be helpful for us to analyze image feature and engraftment, such as transferring color between images. By analogy with Taylor expansion, we designed the image color transfer algorithm by the first and second-order information. The luminance histogram represents the first-order information of image, and the co-occurrence matrix represents the second-order information of image. Some results illustrate our algorithm is effective. In our study, each polynomial in our analogy Taylor expansion of images is considered as one of image features, which makes us re-understand images and its features. It provided us a cue that the features of image, such as color, texture, dimension, time series, would be not isolated but mutual relational based on image expansion.