In the US, colorectal cancer is the second leading cause of cancer-related deaths behind lung cancer, causing about 49,000 annual deaths. Colonoscopy is currently the gold standard procedure for colorectal cancer screening. However, recent data suggest that there is a significant (4-12%) miss-rate for the detection of even large polyps and cancers. To address this, we have been investigating an „automated feedback system‟ which measures quality of colonoscopy automatically by analyzing colonoscopy video frames in order to assist the endoscopist to improve the quality of the actual procedure being performed. One of the fundamental steps analyzing colonoscopy video frames for the automated quality feedback system is to distinguish non-informative frames from informative ones. Most methods to detect and classify these non-informative frames are based on the hand-engineered features. However, it is very tedious to design optimal hand-engineered features. In this paper, we explore the effectiveness of Convolutional Neural Network (CNN) to detect and classify these non-informative frames. The experimental results show that the proposed approaches are promising.