LASER FLUORESCENCE ENDOSCOPE WITH COMPUTER-ASSISTED IMAGE PROCESSING FOR DETECTION OF CANCER
具有图象处理功能的激光荧光内窥系统的研究

The imaging system of LFE (Laser Fluorescence Endoscope) is a new useful tool for detection of early cancer, but a few false positive or negative results during human examination may appear in the current LFE system developed. The surrounding normal tissue near tumor emits auto fluorescence spectrum which peaks in blue when excited by violet laser. This emission greatly reduces or eliminates fluorescence intensity contrast between a tumor and its surrounding. In addition, the fluorescence intensity is also influenced by many other factors, such as non-uniform illumination,tissue size,target color and so on. All these may cause mistaken results of diagnosis. The purpose of this study is to improve diagnostic accuracy. We have developed a new technique based on computer assisted image processing. With the new imaging system it has been demonstrated employing an animal lung model. Our test results show that this new LFE system can significantly enhance the image contrast and reduce false positive and negative results for cancer diagnosis.