3D medical image visualization and VE model to determine the pathology zone of tumor evidence-based using some numerical methods and simulation

Abstract

This paper presents some integrated mathematical modeling and simulation for visualizing a 3D medical image and estimating the volume of tumor growth. Thus, these two indicators will determine the pathology zone and to provide revised evidence-based on tumor histology, location, growth and the treatment effect. There are three phases of modeling and simulation for volume visualization of the 3D tumor. The first phase is converting from 2D signal images to 2D digital images based on edge detection of the tumor. Geodesic Active Contour (GAC) model based on additive operator splitting (AOS) will be used to detect the contour line of a brain tumor on 2D images. The second phase is pre-constructing of 3D digital image from the 2D images by applying two numerical models such as an image manifold model (IM) and volume estimation model (VE). The third phase is implementing the numerical simulation and visualizing the 3D medical image on a hardware and software computational platform. The numerical comparison of VE and IM will be investigate using some performance measurements and interpretation in terms of VE, RMSE, run time and computational complexity cost. In this case study, the medical image is based on a set of 2D MRI brain tumor images from Kubang Krian Hospital Malaysia (HKK). The numerical results will determine the pathology zone and to provide revised evidence-based on tumor informatics. As a conclusion, this paper proof an alternative numerical model is superior to construct and visual the 3D medical images. Thus, volumetric image estimation from the 2D image and extended to a 3D volume image is essential for accurate evaluation of the high resolution 3D medical images.