BRAIN METASTASES DETECTION FROM MAGNET RESONANCE IMAGES

Abstract:

Detection of brain metastases in patients with undiagnosed primary cancer is unusual but still an
existing phenomenon. In these cases, identifying the cancer site of origin is non-feasible by visual
examination of magnetic resonance (MR) images. Recently, radiomics has been proposed to analyze
differences among classes of visually imperceptible imaging characteristics. Support vector machine
(SVM) and k-nearest neighbors (k-NN) classifiers were implemented to evaluate the classification
performance. The influence of gray-level quantization for computation of texture features was also
examined. The best classification (AUC = 0.953 ± 0.061), evaluated with nested cross-validation, was
obtained using the SVM classifier with two texture features derived from the 16 gray-level quantization
co-occurrence matrix.

Proposed system:

We propose a method using contrast-enhanced magnetic resonance(MR) images of brain metastases

from lung and breast cancer to establish a model able to differentiate both classes by means of Texture
analysis and classification techniques. In proposed method using MR images of brain metastases from
lung and breast cancer are detected using edge detection , where detected region boundaries are
extracted using Boundary extraction technique. Region grow technique and pixel masking is used to
extract the features. Based on the features obtained , using Artificial Neural Network Classifier region
properties of metastases .

Block Diagram:

INPUT PRE- Connected Boundary

IMAGE PROCESSING Edge Detection Component Extraction
Analysis

Region Grow
Trained
Technique
dataset

Feature-Based
RESULT Pixel Making
ANN
Classification

Hardware & Software Requirements:

Hardware:

1. 20 Gb Hard disc space

2. 2Gb RAM

Software :

1. MATLAB R2013a

System Requirement:

1. Windows 7 (32 or 64 bit)

2. Windows 8 (32 or 64 bit)