A Benchmark for Endoluminal Scene Segmentation of Colonoscopy Images
This work addresses colorectal cancer screening challenges for clinicians by providing a new benchmark, but it is incremental as it builds on existing methods with improved performance.
The paper tackles the problem of polyp miss-rates and visual assessment limitations in colonoscopy by introducing an extended benchmark for colonoscopy image segmentation, and it reports that training standard fully convolutional networks on this dataset significantly outperforms prior results without post-processing.
Colorectal cancer (CRC) is the third cause of cancer death worldwide. Currently, the standard approach to reduce CRC-related mortality is to perform regular screening in search for polyps and colonoscopy is the screening tool of choice. The main limitations of this screening procedure are polyp miss-rate and inability to perform visual assessment of polyp malignancy. These drawbacks can be reduced by designing Decision Support Systems (DSS) aiming to help clinicians in the different stages of the procedure by providing endoluminal scene segmentation. Thus, in this paper, we introduce an extended benchmark of colonoscopy image, with the hope of establishing a new strong benchmark for colonoscopy image analysis research. We provide new baselines on this dataset by training standard fully convolutional networks (FCN) for semantic segmentation and significantly outperforming, without any further post-processing, prior results in endoluminal scene segmentation.