TY - JOUR
T1 - Fractal Dimension Estimation for Developing Pathological Brain Detection System Based on Minkowski-Bouligand Method
AU - Zhang, Yu Dong
AU - Chen, Xian Qing
AU - Zhan, Tian Ming
AU - Jiao, Zhu Qing
AU - Sun, Yi
AU - Chen, Zhi Min
AU - Yao, Yu
AU - Fang, Lan Ting
AU - Lv, Yi Ding
AU - Wang, Shui Hua
N1 - Publisher Copyright:
© 2016 IEEE.
PY - 2016
Y1 - 2016
N2 - It is of enormous significance to detect abnormal brains automatically. This paper develops an efficient pathological brain detection system based on the artificial intelligence method. We first extract brain edges by a Canny edge detector. Next, we estimated the fractal dimension using box counting method with grid sizes of 1, 2, 4, 8, and 16, respectively. Afterward, we employed the single-hidden layer feedforward neural network. Finally, we proposed an improved particle swarm optimization based on three-segment particle representation, time-varying acceleration coefficient, and chaos theory. This three-segment particle representation encodes the weights, biases, and number of hidden neuron. The statistical analysis showed the proposed method achieves the detection accuracies of 100%, 98.19%, and 98.08% over three benchmark data sets. Our method costs merely 0.1984 s to predict one image. Our performance is superior to the 11 state-of-the-art approaches.
AB - It is of enormous significance to detect abnormal brains automatically. This paper develops an efficient pathological brain detection system based on the artificial intelligence method. We first extract brain edges by a Canny edge detector. Next, we estimated the fractal dimension using box counting method with grid sizes of 1, 2, 4, 8, and 16, respectively. Afterward, we employed the single-hidden layer feedforward neural network. Finally, we proposed an improved particle swarm optimization based on three-segment particle representation, time-varying acceleration coefficient, and chaos theory. This three-segment particle representation encodes the weights, biases, and number of hidden neuron. The statistical analysis showed the proposed method achieves the detection accuracies of 100%, 98.19%, and 98.08% over three benchmark data sets. Our method costs merely 0.1984 s to predict one image. Our performance is superior to the 11 state-of-the-art approaches.
KW - K-fold cross validation
KW - Minkowski Bouligand dimension
KW - artificial bee colony
KW - genetic algorithm
KW - logistic map
KW - number of hidden neuron
UR - http://www.scopus.com/inward/record.url?scp=84994432462&partnerID=8YFLogxK
U2 - 10.1109/ACCESS.2016.2611530
DO - 10.1109/ACCESS.2016.2611530
M3 - Article
AN - SCOPUS:84994432462
SN - 2169-3536
VL - 4
SP - 5937
EP - 5947
JO - IEEE Access
JF - IEEE Access
M1 - 7572925
ER -