Parameterization of BFO Algorithm for the Improved Functionality of MFKM Technique for Better Pathological Identification in Brain MR Image

Intensity inhomogeneity, high level of noise, partial volume effect and poor image contrast are the major artefacts in medical image segmentation. Any of these artefacts might lead to unclear boundaries of tissues, hence the segmentation of tissues in the MR brain image cannot be determined with high accuracy, and this would be a problem to the radiologists to diagnose or to start the treatment because of the lack of facility to operate over the brain in in-vivo condition. This makes the radiologist and surgeons/experts to take time to come for the conclusion on pathology of a particular patient. So, the radiologists and experts need to give more exertion when this condition is applied for many patients at a day, to diagnose and to start treatment. To make this effortless to them, also for accurate diagnosis, this research paper provides an robust algorithm using the Modified Fuzzy K-Means (MFKM) and Bacteria Foraging Optimization (BFO) algorithm, which segments the abnormal tissues among the normal tissues from MR brain images with high accuracy. The accuracy of the Improved MFKM (IMFKM) algorithm is obtained in terms of Sensitivity and Specificity, and the proposed algorithm proves better segmentation results than the other conventional algorithms.

Download Full-text

Modified Kernel Based Fuzzy Clustering for MR Brain Image Segmentation using Deep Learning

International Journal of Engineering and Advanced Technology - Regular Issue ◽

10.35940/ijeat.f8790.088619 ◽

2019 ◽

Vol 8 (6) ◽

pp. 2881-2887

Keyword(s):

Fuzzy Clustering ◽

Bias Field ◽

Mr Images ◽

Brain Images ◽

Normal Tissues ◽

Segmentation Algorithms ◽

Segmentation Approach ◽

Mr Brain Images ◽

Mr Brain ◽

Abnormal Tissue

The segmentation procedure might cause error in diagnosing MR images due to the artifacts and noises that exist in it. This may lead to misclassifying normal tissue as abnormal tissue. In addition, it is also required to model the ontogenesis of a tumour, as they propagate at distinctive rates in contrast to their surroundings. Hence, it is still a challenging task to segment MR brain images due to possible noise presence, bias field and impact of partial volume. This article presents an efficient approach for segmenting MR brain images using a modified kernel based fuzzy clustering (MKFC) algorithm. In addition, this approach computes the weight of each picture element based on the local mutation coefficient (LMC). The proposed system would reflexively group normal tissues like white matter (WM), gray matter (GM) and cerebrospinal fluid (CSF) respectively, from abnormal tissue, such as a tumour region, in MR brain images. Simulation outcomes have shown that the performance of the proposed segmentation approach is superior to the existing segmentation algorithms in terms of both ocular and quantitative analysis

Download Full-text

A Novel Histogram Region Merging Based Multithreshold Segmentation Algorithm for MR Brain Images

International Journal of Biomedical Imaging ◽

10.1155/2017/9759414 ◽

2017 ◽

Vol 2017 ◽

pp. 1-6 ◽

Cited By ~ 1

Author(s):

Siyan Liu ◽

Xuanjing Shen ◽

Yuncong Feng ◽

Haipeng Chen

Keyword(s):

Time Complexity ◽

High Accuracy ◽

Segmentation Algorithm ◽

Gray Level ◽

Original Image ◽

Region Merging ◽

Brain Images ◽

Running Time ◽

Mr Brain Images ◽

Mr Brain

Multithreshold segmentation algorithm is time-consuming, and the time complexity will increase exponentially with the increase of thresholds. In order to reduce the time complexity, a novel multithreshold segmentation algorithm is proposed in this paper. First, all gray levels are used as thresholds, so the histogram of the original image is divided into 256 small regions, and each region corresponds to one gray level. Then, two adjacent regions are merged in each iteration by a new designed scheme, and a threshold is removed each time. To improve the accuracy of the merger operation, variance and probability are used as energy. No matter how many the thresholds are, the time complexity of the algorithm is stable at O(L). Finally, the experiment is conducted on many MR brain images to verify the performance of the proposed algorithm. Experiment results show that our method can reduce the running time effectively and obtain segmentation results with high accuracy.

Download Full-text