Abstract
Prediction of patient survival with brain tumor can play an important role in treatment planning and quality of a patient's life. This study investigates the effect of permeability and perfusion parameters on survival rate. This study identifies the most effective parameters on the survival time and then combines these parameters with experts' knowledge to construct a fuzzy-based model for predicting the survival rate of patients with Glioblastoma Multiforme (GBM). 25 treatment-naive GBM patients were studied using Dynamic Contrast Enhanced Magnetic Resonance Imaging. Using the standard Toft-model combined with Model Selection technique and Maximum-Likelihood algorithm, for each patient, the following permeability parameters were estimated: plasma volume (vp), outward and inward transfer constants (ktrans and kep). The estimated parameters were averaged across the tumor volume for each patient. Histogram of each parameter in whole volume of tumor was calculated and a Normal distribution was assigned to each histogram. Mean and standard deviation of each normal distribution were used to calculate the probability of each parameter being effective on survival rate. All the estimated parameters along with age, gender and hemodynamic parameters (relative Cerebral Blood Volume, rCBV and Cerebral Blood Flow, rCBF) were used as independent features for Cox-proportional-hazard regression model and Wald-test for survival analysis and to identify the parameters with most predictive power of survival. This pilot study revealed that the following physiological parameters are the best candidates to construct the fuzzy-based model for prediction of GBM patient survival: Mean and standard deviation of kep and extra-vascular extra-cellular volume (ve = ktrans/kep), vp, patient age, and rCBV. Compared to the other conventional predictors, the proposed model in this study showed a superior predictive power (∼ 93%) which is due to the fact that it can benefit from experimental data and a set of rules that can be constructed based on the knowledge of experts.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
