A comparative study on polyp classification using convolutional neural networks

Abstract

Colorectal cancer is the third most common cancer diagnosed in both men and women in the United States. Most colorectal cancers start as a growth on the inner lining of the colon or rectum, called 'polyp'. Not all polyps are cancerous, but some can develop into cancer. Early detection and recognition of the type of polyps is critical to prevent cancer and change outcomes. However, visual classification of polyps is challenging due to varying illumination conditions of endoscopy, variant texture, appearance, and overlapping morphology between polyps. More importantly, evaluation of polyp patterns by gastroenterologists is subjective leading to a poor agreement among observers. Deep convolutional neural networks have proven very successful in object classification across various object categories. In this work, we compare the performance of the state-of-the-art general object classification models for polyp classification. We trained a total of six CNN models end-to-end using a dataset of 157 video sequences composed of two types of polyps: hyperplastic and adenomatous. Our results demonstrate that the state-of-the-art CNN models can successfully classify polyps with an accuracy comparable or better than reported among gastroenterologists. The results of this study can guide future research in polyp classification.

Intelligent recognition of colorectal cancer combining application of computer-assisted diagnosis with deep learning approaches

“…The block diagram for intelligent CAD [8] for detection of colorectal is shown in Figure 4. The colonoscopy technique delivers multiple image numbers in the form of image data sets [9]. Remember that these source image data sets should be of the highest quality possible.…”

Colonoscopy polyp detection and classification: Dataset creation and comparative evaluations

“…Another model uses a slightly different approach using 3 different extracted features, color and texture clues, temporal features, and shape to feed an ensemble of 3 CNN models [ 44 ]. Deep learning models have been widely applied to medical problems like anatomical classification, lesion detection, and polyp detection and classification in colonoscopy [ 45 – 50 ]. In [ 45 ], Six classical image classification models have been compared to determine the categories of detected polyps.…”

“…Deep learning models have been widely applied to medical problems like anatomical classification, lesion detection, and polyp detection and classification in colonoscopy [ 45 – 50 ]. In [ 45 ], Six classical image classification models have been compared to determine the categories of detected polyps. It assumes all polyps have been detected and cropped out from the original sequences.…”

Artificial intelligence for the early detection of colorectal cancer: A comprehensive review of its advantages and misconceptions

“…There are two well-defined cases: Images with polyps (positive cases) and images without polyps (negative cases). In both cases, some authors[ 15 , 27 , 28 ] define a true positive (TP) when the algorithm output finds the correct region of the polyp (detection) or labels the image as a polyp (classification). In the case of detection, only one TP is considered per polyp, avoiding over-detection.…”