A modified fuzzy C-means method for segmenting MR images using non-local information

Feng, Yuan; Guo, Hao; Zhang, Hongmiao; Li, Chungang; Sun, Lining; Mutic, Sasa; Ji, Songbai; Hu, Yanle

doi:10.3233/thc-161208

Cited by 9 publications

(9 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Conventionally, the distance between each gray‐level pixel‐valued scalar

x_{i}

and the centroid value scalar

v_{k}

is defined by Euclid distance. In a previous study, we found that a Hausdorff distance based FCM (HFCM) is useful to segment MR images . In HFCM, Euclid distance was replaced with Hausdorff distance:

false‖ p_{i} - p_{j} ‖_{H} = max \{\underset{p_{i} \in A}{falsesup} \underset{p_{j} \in B}{falseinf} false| p_{j} - p_{i} false|, \underset{p_{j} \in B}{falsesup} \underset{p_{i} \in A}{falseinf} false| p_{i} - p_{j} false|\},

where

A = \{p_{i} false| p_{i} \in A\}

and

B = \{p_{j} false| p_{j} \in B\}

are two non‐empty pixel subsets with pixel

p_{i}

p_{j}

located at the center [Fig.…”

Section: Methodsmentioning

confidence: 99%

“…In a previous study, we found that a Hausdorff distance based FCM (HFCM) is useful to segment MR images. 29 In HFCM, Euclid distance was replaced with Hausdorff distance:…”

Section: A Euclid and Hausdorff Distance Based Fcmmentioning

confidence: 99%

“…Many modifications and improvements of the FCM method have been proposed, for example, relational FCM, spatial models, probability based FCM, and a neutrosophic c‐means (NCM) clustering algorithm . It has been revealed that different distance measure based FCM can increase the performance and robustness of the classical FCM algorithm . Notably, Chen and Zhang (2004) are among the first to introduce a kernel‐induced distance measure to FCM and showed the algorithm was robust to noise.…”

Section: Introductionmentioning

confidence: 99%

“…16 It has been revealed that different distance measure based FCM can increase the performance and robustness of the classical FCM algorithm. [27][28][29][30][31] Notably, Chen and Zhang (2004) are among the first to introduce a kernel-induced distance measure to FCM and showed the algorithm was robust to noise.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

An adaptive Fuzzy C-means method utilizing neighboring information for breast tumor segmentation in ultrasound images

et al. 2017

Self Cite

View full text Add to dashboard Cite

show abstract

“…Conventionally, the distance between each gray‐level pixel‐valued scalar

x_{i}

and the centroid value scalar

v_{k}

is defined by Euclid distance. In a previous study, we found that a Hausdorff distance based FCM (HFCM) is useful to segment MR images . In HFCM, Euclid distance was replaced with Hausdorff distance:

false‖ p_{i} - p_{j} ‖_{H} = max \{\underset{p_{i} \in A}{falsesup} \underset{p_{j} \in B}{falseinf} false| p_{j} - p_{i} false|, \underset{p_{j} \in B}{falsesup} \underset{p_{i} \in A}{falseinf} false| p_{i} - p_{j} false|\},

where

A = \{p_{i} false| p_{i} \in A\}

and

B = \{p_{j} false| p_{j} \in B\}

are two non‐empty pixel subsets with pixel

p_{i}

p_{j}

located at the center [Fig.…”

Section: Methodsmentioning

confidence: 99%

“…In a previous study, we found that a Hausdorff distance based FCM (HFCM) is useful to segment MR images. 29 In HFCM, Euclid distance was replaced with Hausdorff distance:…”

Section: A Euclid and Hausdorff Distance Based Fcmmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

An adaptive Fuzzy C-means method utilizing neighboring information for breast tumor segmentation in ultrasound images

et al. 2017

Self Cite

View full text Add to dashboard Cite

show abstract

“…Enhanced systems approach can improve the ability of the ambiguity of the system dealing with uncertainty, Zadeh [1] ambiguity extent of the problem in the concept of fuzzy sets proposed had mentioned the collection of a fuzzy set extension, given the set membership the degree of blurring value, so that the collection descriptions are more ambiguity, this extension of fuzzy set is called type-2 fuzzy sets. Mendel J M presented two complete systems approach based on fuzzy type-2 fuzzy sets on that type-2 fuzzy systems have broader application prospects [2].…”

Section: Introductionmentioning

confidence: 99%

Application of Fuzzy C-means clustering algorithm in image segmentation

Guo¹

2016

Proceedings of the 2016 4th International Conference on Electrical &Amp; Electronics Engineering and Computer Science (ICEEECS

View full text Add to dashboard Cite

Data mining was a core process step in the whole data processing system. The aim is to use specific data mining algorithms to extract knowledge of interest. The user from the database and represented in a certain way, such as the generation rules. Typically, data mining algorithms can be implemented features include: class concept descriptions, association rules, classification regression, cluster analysis, sequence timing analysis, and isolated point analysis. This paper describes one of the most common algorithms based on fuzzy clustering algorithm, which called FCM objective function. FCM algorithm discussed two types of improvement ideas: First, the introduction of clustering validity function in FCM algorithm process to determine the number of clusters c. Second is the evolutionary computation is introduced into the FCM algorithm. On this basis, an improved FCM algorithm, namely through the introduction of simulated annealing particle swarm algorithm FCM is improved to reduce the impact of isolated points on the cluster center.

show abstract

Quantification of shoulder muscle intramuscular fatty infiltration on T1-weighted MRI: a viable alternative to the Goutallier classification system

et al. 2018

View full text Add to dashboard Cite

show abstract

A modified fuzzy C-means method for segmenting MR images using non-local information

Cited by 9 publications

References 30 publications

An adaptive Fuzzy C-means method utilizing neighboring information for breast tumor segmentation in ultrasound images

An adaptive Fuzzy C-means method utilizing neighboring information for breast tumor segmentation in ultrasound images

Application of Fuzzy C-means clustering algorithm in image segmentation

Quantification of shoulder muscle intramuscular fatty infiltration on T1-weighted MRI: a viable alternative to the Goutallier classification system

Contact Info

Product

Resources

About