Heterogeneous data analysis: Online learning for medical-image-based diagnosis
作者：
Highlights：
• Proposes a method of Online Principal Composite Kernel Feature Analysis.
• Develops Heterogeneous Big Data Associations to improve data quality.
• Compares Online and Office Learning for Computer-Aided Detection of cancer polyps.
• Implements online diagnosis with long-term sequential datasets.
• Evaluates the resulting data quality and computational time.
摘要
•Proposes a method of Online Principal Composite Kernel Feature Analysis.•Develops Heterogeneous Big Data Associations to improve data quality.•Compares Online and Office Learning for Computer-Aided Detection of cancer polyps.•Implements online diagnosis with long-term sequential datasets.•Evaluates the resulting data quality and computational time.
论文关键词：Online learning,Computed tomographic colonography,Heterogeneous data analysis,Kernel feature analysis,Computer-aided detection,Principal composite kernel feature analysis
论文评审过程：Received 21 December 2015, Revised 25 May 2016, Accepted 21 September 2016, Available online 29 September 2016, Version of Record 27 November 2016.
论文官网地址：https://doi.org/10.1016/j.patcog.2016.09.035
