“School of Cognitive”
Back to Papers HomeBack to Papers of School of Cognitive
| Paper IPM / Cognitive / 13439 |
|
||||||
| Abstract: | |||||||
| 1. Introduction Multiple Sclerosis (MS) is autoimmune diseases that immune system is damage the nerves of brain and spinal cord. In MS disease the myelin of axons is distorted and the extracellular liquid is accumulated in these areas. Magnetic resonance imaging (MRI) using mentioned MS characteristics is often used to quantify MS lesions in brain. 2. Materials |
methods
The dataset is a benchmark dataset used in the Medical Image Computing and Computer Aided Intervention Society's (MICCAI's) MS Lesion Segmentation challenge 2008 (data set consist of 8 subjects). In this work, a fully automated adaptive technique has been implemented to detect WM lesions in MR images. Our proposed method is divided into 3 stages: (a) the preprocessing step to reduce noise, which includes skull stripping (using FSL) and bias filed correction (using Fuzzy C-Means algorithm) (b) the WM lesions segmentation step (contrast based); and (c) the post processing step, which includes normal brain tissue segmentation and morphological processing to remove false positives (FPs).
3. Results
In this study after that the WM lesions detected, the false positive (FP) and True positive (TP) index calculated to compare our results with radiologist result. The result shows that we could able to detect WM lesion with more than 784. Conclusion
In this study, we demonstrate that we were able to detect the WM lesions effectively without requiring to standard atlas or training procedures.
Download TeX format |
||||||
| back to top | |||||||
