TY - JOUR
T1 - Comparison of machine learning methods for stationary wavelet entropy-based multiple sclerosis detection
T2 - Decision tree, k -nearest neighbors, and support vector machine
AU - Zhang, Yudong
AU - Lu, Siyuan
AU - Zhou, Xingxing
AU - Yang, Ming
AU - Wu, Lenan
AU - Liu, Bin
AU - Phillips, Preetha
AU - Wang, Shuihua
N1 - Publisher Copyright:
© The Author(s) 2016.
PY - 2016/9/1
Y1 - 2016/9/1
N2 - In order to detect multiple sclerosis (MS) subjects from healthy controls (HCs) in magnetic resonance imaging, we developed a new system based on machine learning. The MS imaging data was downloaded from the eHealth laboratory at the University of Cyprus, and the HC imaging data was scanned in our local hospital with volunteers enrolled from community advertisement. Inter-scan normalization was employed to remove the gray-level difference. We adjust the misclassification costs to alleviate the effect of unbalanced class distribution to the classification performance. We utilized two-level stationary wavelet entropy (SWE) to extract features from brain images. Then, we compared three machine learning based classifiers: the decision tree, k-nearest neighbors (kNN), and support vector machine. The experimental results showed the kNN performed the best among all three classifiers. In addition, the proposed SWE+kNN approach is superior to four state-of-the-art approaches. Our proposed MS detection approach is effective.
AB - In order to detect multiple sclerosis (MS) subjects from healthy controls (HCs) in magnetic resonance imaging, we developed a new system based on machine learning. The MS imaging data was downloaded from the eHealth laboratory at the University of Cyprus, and the HC imaging data was scanned in our local hospital with volunteers enrolled from community advertisement. Inter-scan normalization was employed to remove the gray-level difference. We adjust the misclassification costs to alleviate the effect of unbalanced class distribution to the classification performance. We utilized two-level stationary wavelet entropy (SWE) to extract features from brain images. Then, we compared three machine learning based classifiers: the decision tree, k-nearest neighbors (kNN), and support vector machine. The experimental results showed the kNN performed the best among all three classifiers. In addition, the proposed SWE+kNN approach is superior to four state-of-the-art approaches. Our proposed MS detection approach is effective.
KW - Multiple sclerosis
KW - decision tree
KW - k -nearest neighbors
KW - machine learning
KW - stationary wavelet entropy
KW - support vector machine
UR - http://www.scopus.com/inward/record.url?scp=84988736270&partnerID=8YFLogxK
U2 - 10.1177/0037549716666962
DO - 10.1177/0037549716666962
M3 - Article
AN - SCOPUS:84988736270
SN - 0037-5497
VL - 92
SP - 861
EP - 871
JO - SIMULATION
JF - SIMULATION
IS - 9
ER -