Deep Learning Models for Colorectal Polyps

Colorectal cancer is one of the main causes incident cases and deaths worldwide. Undetected colon polyps, be them benign or malignant, lead to late diagnosis colorectal cancer. Computer aided devices have helped decrease polyp miss rate. The application deep learning algorithms techniques has escalated during this last decade. Many scientific studies are published detect, localize, classify polyps. We present here a brief review latest studies. compare accuracy these with our results obtained from training testing three independent datasets using convolutional neural network autoencoder model. A train, validate test split was performed for each dataset, 75%, 15%, respectively. An 0.937 achieved CVC-ColonDB, 0.951 CVC-ClinicDB, 0.967 ETIS-LaribPolypDB. Our suggest slight improvements compared used date.