In order to achieve the automatic identification of liver cancer cells in the blood, the present study adopted a principal component analysis (PCA) and back propagation (BP) algorithm of feedforward neural networks to identify white blood cells and red blood cells in mice and human liver cancer cells, HepG2. The present paper shows the process in which PCA was carried out after obtaining spectral data by fiber confocal back-scattering spectrograph, selecting the first two principal components as spectral features, and establishing a neural network pattern recognition model with two input layer nodes, eleven hidden layer nodes and three output nodes. In order to verify whether the model would give accurate identification of cells, we chose 195 object data to train the model with 150 sets of data as training set and 45 sets as test set. According to the results, the overall recognition accuracy of the three cells was above 90% with the average relative deviation only 4.36%. The results showed that PCA+BP algorithm could automatically identify liver cancer cells from erythrocyte and white blood cells, which will provide a useful tool for the study of metastasis and biological metabolism characteristics of liver cancer.
In photoacoustic imaging the ultrasonic signals are usually detected by contacting transducers. For some applications, contact with the tissue should be avoided, e.g. in those of brain functional imaging. As alternatives to contacting transducers interferometric techniques can be used to acquire photoacoustic signals remotely. Here, a system for non-contact photoacoustic tomography imaging (NCPAT) has been established. This approach enables NCPAT not to exceed laser exposure safety limits. The stimulated source of NCPAT utilized a laser with center wavelength of 532 nm and output intensity of 17.5 mJ/cm2, and a laser heterodyne interferometry was used to receive the photoacoustic signals. The NCPAT was used to implement on a rotational imaging geometry for photoacoustic tomography with a real-tissue phantom. The photoacoustic imaging was obtained by applying a reconstruction algorithm to the data acquired for NCPAT. Experiments results showed that the NCPAT system with detection 15 dB bandwidth of 2.25 MHz could resolve spherical optical inclusions with dimension of 500 μm and multi-layered structure with optical contrast in strongly scattering medium. The method could expand the scope of photoacoustic and ultrasonic technology to in-vivo biomedical applications where contact is impractical.