Infrared Spectroscopy help diagnose cancer

A genetic algorithm developed by British scientists to assist in the use of infrared spectroscopy (IR) in the diagnosis of cancer. IR spectroscopy is a non-destructive method in the analysis of cells, tissues and liquid gas has been used to detect different diseases and levels of malignancies of the prostate infection, cervical and colon tissues. However, some methods for processing the cells before the analysis is neither standard through which different laboratories, which could make the results difficult to interpret by non-specialty in the hospital itself.





The collaboration led by Peter Gardner at the University of Manchester have examined IR signs of prostate cancer using cells from the same source in the body. By transforming the cells are biologically and chemically, the researchers mimicked tumor progression and found that the differences between the signs of the tumor cells with different behavior is very difficult. Gardner developed a genetic algortma - a computer program that studied and optimized solution - to process data from the IR before the cells were analyzed. They use 50 independent genetic algorithm are employed to optimize the number of spectra classified correctly. With this method, they can separate families of related cells are close with a high level of accuracy. Max Diem, an expert on the spectral diagnosis of Northeastern University in Boston, USA, said the paper is "an application that pioneered the technology developed recently, and outlines and power lines discriminate cancer cells from different raid levels by spectral methods'. 'The aim is to identify some tumors are likely to spread, "explained Gardner, and added that with further work, identifying the signs of IR would be an immeasurable price that can be used to aid medical decisions regarding treatment. 'At this time, it is beyond the proof of principle that there are levels, but not yet ready for a full medical examination,' he concluded.