TY - GEN
T1 - Application of inductive learning in human brain CT image recognition
AU - WANG, Xi-Zhao
AU - LIN, Wei-Xi
N1 - This work is financially supported by National Science Fund of China (60473045) and Hebei Province High Technology project (04213533).
PY - 2007
Y1 - 2007
N2 - This paper presented three ways utilized to do head CT (Computed Tomography) image Computer Aided Diagnosis and their performance comparison, and analysis of the performance comparison. In this present work, we collected 116 normal pieces of head CT image and 96 abnormal ones, and utilized two ways to do feature extraction, and applied the ways of decision tree, RBFNN and outlier detection in classification. At the same time, it was implied in this paper that one question formulation could lead to different question solutions by utilizing different methods. The question formulation is the base of measure which we make use of. For this point of view, the question formulation is more important than question solution. Meanwhile, as future work, we will try to find different formulations of this question by making different question solutions as guidance. So we can achieve more conclusions of the question, and can select optimal solution in greater scope.
AB - This paper presented three ways utilized to do head CT (Computed Tomography) image Computer Aided Diagnosis and their performance comparison, and analysis of the performance comparison. In this present work, we collected 116 normal pieces of head CT image and 96 abnormal ones, and utilized two ways to do feature extraction, and applied the ways of decision tree, RBFNN and outlier detection in classification. At the same time, it was implied in this paper that one question formulation could lead to different question solutions by utilizing different methods. The question formulation is the base of measure which we make use of. For this point of view, the question formulation is more important than question solution. Meanwhile, as future work, we will try to find different formulations of this question by making different question solutions as guidance. So we can achieve more conclusions of the question, and can select optimal solution in greater scope.
KW - CT image
KW - Decision tree
KW - Outlier detection
KW - RBFNN
KW - Symmetry
KW - Texture
UR - http://www.scopus.com/inward/record.url?scp=38049078838&partnerID=8YFLogxK
U2 - 10.1109/ICMLC.2007.4370415
DO - 10.1109/ICMLC.2007.4370415
M3 - Conference paper (refereed)
AN - SCOPUS:38049078838
SN - 9781424409723
T3 - International Conference on Machine Learning and Cybernetics (ICMLC)
SP - 1667
EP - 1671
BT - Proceedings of the Sixth International Conference on Machine Learning and Cybernetics, ICMLC 2007
PB - IEEE
T2 - 6th International Conference on Machine Learning and Cybernetics, ICMLC 2007
Y2 - 19 August 2007 through 22 August 2007
ER -