International Conference on Computer and Communication Engineering (ICCCE 2012), 3-5 July 2012, Kuala Lumpur, Malaysia 978-1-4673-0479-5/12/$31.00 ©2012 IEEE Brain Tumor Quantification Equation: Modeled on Complete Step Response Algorithm Abdulfattah A. Aboaba 1, 2 , Shihab A. Hameed 1 , Othman O. Khalifa 1 , Aisha H. Abdalla 1 1 Department of Electrical & Computer Engineering, International Islamic University Malaysia 2 Computer Engineering Department, University of Maiduguri Nigeria abdulfattahaa@gmail.com., shihab@iiu.edu.my, khalifa@iiu.edu.my, aisha@iiu.edu.my Rahmat Harun 3 , Norzaini Rose Mohd Zain 4 3 Klinik Pakar Neurosurgeri, Hospital Queen Elizabeth, Kota Kinabalu, Sabah Malaysia 4 Radiology Department, Hospital Kuala Lumpur Kuala Lumpur Malaysia rahmat865@yahoo.com, norzainirose@yahoo.co.uk Abstract— In Image Guided neuro-Surgery (IGnS) protocol relating to tumor, the planning stage is the bottleneck where most times are spent reconstructing the slices in order to; quantify the tumor, get the tumor shape and location relative to adjacent cells, and determine best incursion route among others. This time consuming assignment is handled by a surgeon using any of the standardized IGnS software. It has been observed that the approach taken to quantify tumor in those software are simply replicating the surgeons’ experience-based brain tumor quantification technique fashionable in the pre-imaging era. The result is a quantification method that is time consuming, and at bests an approximation. What is presented here is a novel brain tumor quantification method based on step response algorithm utilizing a model which itself was based on step response model resulting in smart and rapid quantification of brain tumor volume. Keywords- Image Guided neuro-Surgery; Brain Tumor Quantification; Step Response Model; Smart quantification. I. INTRODUCTION One of the core principles in maintenance engineering is that the time to detect fault and restore the system back to good working condition should be as small as possible to reduce down-time. When the whole IGnS protocol is observed, the time spent at the planning stage is phenomenal [1] especially when the surgeon is new or inexperienced in IGS protocol. Many approaches have been suggested for reducing the time spent on Image Guided Surgery (IGS) as a whole [2][3][4][5]. Moreover several works has equally been done that were aimed at getting a growth model or tumor treatment [6][7][8]. However, we observed that since one of the work at that stage is to know the size and shape of tumor, for tumor related operation, we thus observed that, having a good model of brain tumor growth pattern as in [9], and a mathematical equation based on the model will help in quickly determining the size of tumor in the patients’ brain without having to spent time at the planning stage. The rest of the paper is arranged thus: Brain tumor growth model was discussed in section two, section three is on the derivation of mathematical equation for tumor quantification, application of the equation as a proof of concept and design of validation technique was the subject of section four while section five discussed the result and gave a summary of the work. Conclusion was draw in section six together with our future work. II. BRAIN TUMOR GROWTH MODEL The work in [9] proposed that the general pattern of growth of brain tumor on slice by slice basis is governed by equation (1): Where and are and respectively, h is slice thickness, and is the tumor growth and decay model called behavioural function. It is expanded as: Where stand for tumor area at any point within slice thickness (h), tumor area at the lower surface of slice, and tumor area at the upper surface of slice assuming that the tumor area at the lower surface is less than that of the upper surface. The work went further to decompose into a function of (nabla) and (rho) as in: ………(3) in which nabla is a constant called behavioural factor and static part of , it is a kind of generic curve for tumor growth and decay whose numerical value was experimentally fixed at 1.1 as in figure 1, and rho is the dynamic part of which moves nabla within the quadrant. This resulted into an exponential model for finding the instantaneous area of tumor in a slice as in equation (4) [last work] 988