City of Hope, Department of Surgery, Duarte, California, United States.
National Cancer Institute, National Institutes of Health Campus, Department of Surgery, Bethesda, Ma, United States.
J Biomed Opt. 2021 Jan;26(1). doi: 10.1117/1.JBO.26.1.015001.
Colorectal cancer incidence has decreased largely due to detection and removal of polyps. Computer-aided diagnosis development may improve on polyp detection and discrimination.
To advance detection and discrimination using currently available commercial colonoscopy systems, we developed a deep neural network (DNN) separating the color channels from images acquired under narrow-band imaging (NBI) and white-light endoscopy (WLE).
Images of normal colon mucosa and polyps from colonoscopies were studied. Each color image was extracted based on the color channel: red/green/blue. A multilayer DNN was trained using one-channel, two-channel, and full-color images. The trained DNN was then tested for performance in detection of polyps.
The DNN performed better using full-colored NBI over WLE images in the detection of polyps. Furthermore, the DNN performed better using the two-channel red + green images when compared to full-color WLE images.
The separation of color channels from full-color NBI and WLE images taken from commercially available colonoscopes may improve the ability of the DNN to detect and discriminate polyps. Further studies are needed to better determine the color channels and combination of channels to include and exclude in DNN development for clinical use.
结直肠癌的发病率已经大大降低，这主要归因于息肉的检测和切除。计算机辅助诊断的发展可能会提高息肉的检测和鉴别能力。
为了利用现有的商用结肠镜系统提高检测和鉴别能力，我们开发了一种深度神经网络（DNN），用于分离窄带成像（NBI）和白光内镜（WLE）下采集的图像的颜色通道。
研究了结肠镜检查中正常结肠黏膜和息肉的图像。根据颜色通道（红/绿/蓝）提取每个彩色图像。使用单通道、双通道和全彩色图像对多层 DNN 进行训练。然后，对训练好的 DNN 进行检测息肉的性能测试。
与 WLE 图像相比，DNN 在用全彩色 NBI 检测息肉时表现更好。此外，与全彩色 WLE 图像相比，DNN 在使用双通道红色+绿色图像时表现更好。
从商用结肠镜采集的全彩色 NBI 和 WLE 图像中分离颜色通道可能会提高 DNN 检测和鉴别息肉的能力。需要进一步的研究来更好地确定用于临床应用的 DNN 开发中包含和排除的颜色通道和通道组合。
