A New Data Processing Approach Research to Auto-fluorescence Spectrogram for Colorectal Carcinoma

Data classification is an important data mining role in biomedicine. This paper proposes a method to analyze colorectal carcinoma auto-fluorescence spectrogram data based on counting kNN algorithm after analyzing the characteristics of biomedicine data. Though counting kNN algorithm for classification is simple and effective, it doesn´t deal with biomedicine data well. After analyzing the algorithm performance, a novel counting kNN algorithm by index tree is presented. The new method improves the efficiency by using a tree structure index with the same accuracy. Experiments show that this method outperforms the distance-based voting kNN for accuracy, and ckNN for efficiency.