Frontiers in Biomedical Technologies 2016. 3(1-2):8-19.

An Investigation into the Performance of Adaptive Neuro-Fuzzy Inference System for Brain Tumor Delineation Using ExpectationMaximization Cluster Method; a Feasibility Study
Leila Ghorbanzadeh, Ahmad Esmaili Torshabi




Purpose: In brain MRI analysis, image segmentation is commonly used for measuring and visualizing the brain’s anatomical structures, for analyzing brain changes, delineating pathological regions, and image-guided interventions. Since manual segmentation is time-consuming and prone to variable sort of errors, which makes automatic techniques more demanding.

Method: This paper describes a framework for automatic segmentation of both normal and abnormal anatomy from medical images based on adaptive neuro-fuzzy inference system (ANFIS) which is applicable to different types of tumors. The segmentation framework is comprised of five stages: first, Median filter is applied to remove or reduce the noise of images; second, it is followed by EM clustering to segment it into different parts with variousintensities, to be used for feature extraction in the next step. At the fourth stage, extracted features besides ground truth are used as ANFIS training dataset. Fifth and the last, fordetected abnormal sections either edema or tumor core, level set is adopted for a precise detection of abnormal tissues.

Results: This method was applied for 15 High-Grade (HG) and 15 Low-Grade (LG) simulated brain tumor images. Proposed model provided satisfactory outcomes which for the segmentation of whole tumor including both edema and tumor core, Dice index recorded 0.936±0.04 and 0.921±0.02 for HG and LG dataset respectively; however, those of tumor core were 0.899±0/04 and 0.902±0.05 in the mentioned groups.

Conclusion: The results of this study prove fuzzy inference systems and neural networks potential applications in clinical image analysis and tumor evaluation for brain cancers. 

A Method for Brain Tumor Delineation Using Adaptive Neuro-Fuzzy Inference System in Combination with Expectation-Maximization Clustering; a Feasibility Study

Full Text:



  • There are currently no refbacks.

Creative Commons Attribution-NonCommercial 3.0

This work is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported License which allows users to read, copy, distribute and make derivative works for non-commercial purposes from the material, as long as the author of the original work is cited properly.