Gastric Cancer Detection by Mass Spectrum Feature Extraction Based on ICA
SELDI-TOF is a powerful analyzing technique in Gastric Mass Spectrometry based Gastric Cancer (GC) early detection. However, extracting features from the vast amount of spectrum data has become the bottleneck preventing more accurate classification. Independent component analysis (ICA) has proved useful in decomposing observed signals into independent components. We propose it for dimension reduction and protein mass spectra data feature extraction. Data processed by ICA performed better compared to data extracted by software. This approach is a promising tool in preparing high dimensional data for classification.
Mass spectrum gastric cancer ICA feature extraction classification
Jizhe Wang Jun Meng Fuming Qiu Jian Huang
Institute of Electrical Automation, College of Electrical Engineering, Zhejiang University Hangzhou, Department of Oncology, Second Affiliated Hospital, Zhejiang University School of Medicine Hangzhou,
国际会议
成都
英文
28-31
2011-01-14（万方平台首次上网日期，不代表论文的发表时间）