Coriat R, Chryssostalis A, Zeitoun J D, Deyra J, Gaudric M, Prat F, Chaussade S
Service de Gastroentérologie, CHU Cochin, 27 Rue du Faubourg-Saint-Jacques, 75014 Paris, France.
Gastroenterol Clin Biol. 2008 Apr;32(4):363-9. doi: 10.1016/j.gcb.2007.11.013. Epub 2008 Mar 19.
A newly developed computed virtual chromoendoscopy system, Fuji Intelligent Color Enhancement (FICE) technology, decomposes images by wavelength, then directly produces reconstructed images with enhanced mucosal surface contrast. The aims of the present study were to determine the quality of information provided by computed virtual chromoendoscopy for detecting gastrointestinal polyps and to identify the best channel setting for imaging.
Thirty-one upper endoscopy procedures were performed in 31 patients using Fujinon 1.3-million-pixel endoscopes with zoom. A polyp was diagnosed in 45% of the patients. Six experienced endoscopists, who had never used the computed virtual chromoendoscopy system before, analysed the 31 endoscopy reports. Each endoscopy report included 11 images (one conventional image and one image for each FICE channel). The endoscopists used a 10-cm analog visual scale to determine the three best FICE channels, and to evaluate the quality and pit pattern of the chromoendoscopy images.
Channel 4 (red: 520; green: 500; blue: 405) was considered the best channel in 39.7% of the reports (p<0.0001) and was among the best three in 77%. For 94.1% of the reports, the best FICE channel image was considered superior to the conventional image (p<0.0001). Median pit pattern score was 9.43 for computed virtual chromoendoscopy and 7.08 for conventional endoscopy (p<0.001).
Fuji Intelligent Color Enhancement (FICE) channel 4 images were significantly better than conventional images. Computed virtual chromoendoscopy enabled better analysis of the pit pattern and the normal-pathological mucosal junction. Computed virtual chromoendoscopy can be used to identify gastroduodenal polyps and to assist in complete polypectomy.
一种新开发的计算机虚拟染色内镜系统，富士智能色彩增强（FICE）技术，通过波长分解图像，然后直接生成具有增强黏膜表面对比度的重建图像。本研究的目的是确定计算机虚拟染色内镜在检测胃肠道息肉方面提供的信息质量，并确定最佳的成像通道设置。
使用具有变焦功能的富士能130万像素内镜对31例患者进行了31次上消化道内镜检查。45%的患者被诊断出有息肉。6名经验丰富但此前从未使用过计算机虚拟染色内镜系统的内镜医师分析了这31份内镜检查报告。每份内镜检查报告包含11张图像（一张传统图像和每个FICE通道的一张图像）。内镜医师使用10厘米的模拟视觉量表来确定三个最佳的FICE通道，并评估染色内镜图像的质量和凹陷模式。
在39.7%的报告中，通道4（红色：520；绿色：500；蓝色：405）被认为是最佳通道（p<0.0001），在77%的报告中该通道位列最佳三个通道之中。在94.1%的报告中，最佳的FICE通道图像被认为优于传统图像（p<0.0001）。计算机虚拟染色内镜的凹陷模式评分中位数为9.43，传统内镜检查为7.08（p<0.001）。
富士智能色彩增强（FICE）通道4的图像明显优于传统图像。计算机虚拟染色内镜能够更好地分析凹陷模式和正常-病理黏膜交界处。计算机虚拟染色内镜可用于识别胃十二指肠息肉并协助进行完整的息肉切除术。
