Data on MRI brain lesion segmentation using K-means and Gaussian Mixture Model-Expectation Maximization

Ju Qiao; Xuezhu Cai; Qian Xiao; Zhengxi Chen; Praveen Kulkarni; Craig Ferris; Sagar Kamarthi; Srinivas Sridhar

doi:10.1016/j.dib.2019.104628

Data on MRI brain lesion segmentation using K-means and Gaussian Mixture Model-Expectation Maximization

Data Brief. 2019 Oct 10:27:104628. doi: 10.1016/j.dib.2019.104628. eCollection 2019 Dec.

Authors

Ju Qiao¹, Xuezhu Cai², Qian Xiao³, Zhengxi Chen⁴, Praveen Kulkarni⁵, Craig Ferris⁵, Sagar Kamarthi¹, Srinivas Sridhar⁶

Affiliations

¹ Department of Mechanical and Industrial, Northeastern University, Boston, MA, USA.
² Department of Bioengineering, Northeastern University, Boston, MA, USA.
³ Department of Pharmacology, Yale University, New Haven, CT, USA.
⁴ Department of Orthodontics, Shanghai Jiaotong University, Shanghai Ninth People's Hospital, Shanghai, China.
⁵ Department of Psychology, Northeastern University, Boston, MA, USA.
⁶ Department of Physics, Northeastern University, Boston, MA, USA.

Abstract

The data in this article provide details about MRI lesion segmentation using K-means and Gaussian Mixture Model-Expectation Maximization (GMM-EM) algorithms. Both K-means and GMM-EM algorithms can segment lesion area from the rest of brain MRI automatically. The performance metrics (accuracy, sensitivity, specificity, false positive rate, misclassification rate) were estimated for the algorithms and there was no significant difference between K-means and GMM-EM. In addition, lesion size does not affect the accuracy and sensitivity for either method.

Keywords: Ischemic stroke; Lesion; Magnetic resonance image (MRI); Segmentation.