Web proceedings papers


Carlo Ciulla , Ustijana Rechkoska Shikoska , Dijana Capeska Bogatinoska , Filip Risteski and Dimitar Veljanovski


This work studies the human brain affected from tumor related pathologies, specifically: (i) a lesion and (ii) a Glioblastoma multiforme. Applica- tions of Magnetic Resonance Imaging (MRI) post-processing techniques are presented. The techniques perform feature extraction and thus collect complementary and/or additional information from the original MRI. The techniques are: (i) the signal resilient to interpolation, (ii) the intensity-curvature functional, (iii) the intensity-curvature measure, and (iv) the classic- curvature. The four biomedical image processing techniques all descend from the concept of classic-curvature, which can be calculated on the brain images through the summa- tion of all of the second order derivatives of the Hessian of the model function fitted to the image data. The results show that through feature extraction of the original MRI it is possible to collect information about the two brain pathologies herein studied. More specifically, the signal resilient to interpolation provides with the specular inverted MRI, the intensity-curvature functional and the classic-curvature provide with the depth information of: (i) the pathology and (ii) the anatomy of the human brain, and the intensity-curvature measure provides with an additional view on the pathology which is complementary to the view obtained through the original MRI.


classic-curvature, intensity-curvature functional, intensity-curvature measure, model polynomial function, signal resilient to interpolation.