| Laser induced auto-fluorescence (LIAF) spectrum is a new way to diagnose cancer in recent years. It has the merits of high accuracy, fast speed, simple manipulation, non-invasive, non-toxicity and no side-effect compared with other traditional diagnostic techniques.Many basic and clinic research about auto-fluorescence were done in diagnosis of lung cancer in domestic and overseas at present. The research of auto-fluorescence in digestive tract especially in colorectal cancer which is much more complicated is mostly in experimental stage and less in number. There is great difference among all the diagnostic standards in different researching institution. So it is not so perfect and needs more experimental investigation to accumulate data for exploring the rule and establishing reliable and practical diagnostic standards.In this dissertation, the main principles, molecule biological basis and primary research in vivo on LIAF spectrum for early diagnosing in colorectal cancer are studied in detail by combining the basic theory with clinical practice. The research in this dissertation is funded by the National Key Research Project of the Ninth-five-year tackle problems in science and technology (NO.96-901-07-04) and the National Key Research Project of the tenth-five-year tackle problems in science and technology (NO.2001BA706B-15)".The main contents of this dissertation include:PART I Laser induced auto-fluorescence (LIAF) spectrum in distinguishing colorectal cancer tissues from corresponding normal tissuesObjective: To evaluate the feasibility of laser induced auto-fluorescence (LIAF) spectrum in distinguishing colorectal cancer tissues from corresponding normal colorectal tissues and establish an equation to diagnose colorectal cancer. Methods: A Nd:YAG laser (380nm) was used as a simulating light source. An optical multichannel analyzer (OMA) was used to analyze the fluorescence spectrum of colorectal cancer tissues and corresponding normal colorectal tissues in 32 patientsIVin vitro. Then we established an equation for distinguishing the colorectal cancer tissues from corresponding normal colorectal tissues and compared the results with those pathologic results. Results: (1) The LIAF spectrum of colorectal cancer tissues was the same as that of normal tissues with two peaks of spectrum. Both the main and the second peak intensities from cancer tissue were lower than those from normal tissue (P<0.01) . (2) The main and the second peak respectively located at about 475nm and 550nm from both colorectal cancer and normal tissue. The intensities in wave length 640nm of fluorescence spectrum of colorectal cancer tissues were higher than that of the corresponding normal tissue in some patients. (3) We have established the mathematic distinguished equation,W =^(6.184375*X1)2+(1.987699* X2)2 + (2.965413 * X3)2 +0.2, by using the improved neural networks model. The parameters, X1= (I475-I55o)/I640, X2=I475 /I640 and X3= I550 /I640, represent the ratio of the integrated spectral intensities in wave length 475nm, 550nm and 640nm. The sample would be colorectal normal tissue if W>12, and colorectal cancer tissue if W<12. The veracity of colorectal cancer distinguishing from normal colorectal tissue is 94% by testing the neural network model from 100 random samples. Conclusion: The LIAF spectrum can distinguish colorectal cancer tissues from the corresponding normal tissues.PART II Feature of laser induced auto fluorescence (LIAF) spectrum of colorectal cancer and its mechanism Objective: To explore the feature of laser induced auto-fluorescence (LIAF) spectrum in detecting colorectal cancer, and determine the mechanism of the production of the auto-fluorescence. Methods: Tissue auto-fluorescence was excited by a Nd:YAG laser with wave length 380nm. An optical multi-channel analyzer (OMA) was used to analyze the sequential auto-fluorescence spectrum from normal colorectal tissues to adjoining tissues and colorectal cancer tissues in 10 patients in vitro. And we also collected the sequential auto-fluorescence spe...
|
PDF Full Text Request