Principal component analysis approach for biomedical sample identification

Robotic control application on remote surgery has initiated an increasing interest recently as a result of the rapid development of the communication technology and multi-sensory integration. Raman spectroscopy can provide detailed information on molecular composition and it enables the detection of sample pathological changes in a non-destructive manner. It is particularly useful for in vivo tissue analysis. A feasible objective is to create a real-time approach of sample analysis using a Raman spectrometer directly mounted at the end-effector of medical robot to enhance the remote robot surgery. In order to extract intrinsic Raman spectrum, the impact of background spectrum needs to be excluded at first. Signal to noise ratio (SNR) can be improved by filtering techniques and the data normalization can be conducted by standard normal variate (SNV). Principal component analysis (PCA) is proposed for sample identification. PCA is used for dimension reduction so that significant signatures for different types of samples are indicated by dominant eigenvectors from the correspondent covariance matrix. Eventually different principal components are selected for cluster separation. By principal component analysis and control oriented identification, various samples can be distinguished in terns of intrinsic Raman spectrum. In this study, PCA identifies tissues from distinct clusters of different organs. A systematic approach is then formulated for sample identification via Raman spectroscopy.