ABSTRACT
An artificial neural network (ANN) has been used in various clinical research for the prediction and classification of data in cancer disease. Previous research in this direction focused on the correlation between various input parameters such as age, antigen, and size of tumor growth. Recently, laser-induced autofluorescence (LIAF) techniques have been shown to be a useful noninvasive early diagnostic tool for various cancer diseases. We report on a successful application of ANN to in vitro LIAF spectra. We show that classification of tumor samples with ANN can be done with high sensitivity, specificity, and accuracy. Thus a combination of LIAF techniques and ANN can provide a robust method for clinical diagnosis.
Subject(s)
Algorithms , Biomarkers, Tumor/analysis , Colorectal Neoplasms/diagnosis , Colorectal Neoplasms/metabolism , Diagnosis, Computer-Assisted/methods , Lasers , Neural Networks, Computer , Spectrometry, Fluorescence/methods , Colorectal Neoplasms/classification , Humans , Reproducibility of Results , Sensitivity and SpecificityABSTRACT
Laser-induced autofluorescence techniques have the potential to be used for the detection of preinvasive human cancer cells. For colorectal and gastrointestinal cancer cells, the light is introduced in vivo through endoscopic means and the probe tip is brought gently into contact with the tissue under investigation. However, it is often assumed that there is no distance or angular dependence in the intensity of the light collected from the probes. We performed an in vitro experiment in which we showed that there was indeed no angular dependence provided the angle of inclination of the probe to normal incidence is small. However, we find substantial fluctuation in the intensities of peaks for changing distances. These fluctuations can be eliminated by considering the ratio of the intensities from two spectral lines.
