Performance of a deep learning system for automatic diagnosis of protruding lesions in colon capsule endoscopy: a multicentric study

• Type: Other Publications
• Date: 2021

Authors

Saraiva, MJM (Author) Other; Ferreira, J (Author) FEUP; Cardoso, H (Author) Other; Afonso, J (Author) Other; Ribeiro, T (Author) Other; Andrade, P (Author) Other; Parente, MPL (Author) FEUP; Natal, R (Author) Other; Saraiva, MM (Author) Other; Macedo, G (Author) Other

Other information

Authenticus ID: P-00Z-22S

DOI: 10.21203/rs.3.rs-284396/v1

Abstract (EN): <jats:title>Abstract</jats:title> <jats:p>Colon capsule endoscopy (CCE) represents a landmark in minimally invasive exploration of the colonic mucosa for patients with contraindications for a conventional colonoscopy, or for whom the latter exam is unwanted or unfeasible. Colorectal neoplasia is the most common lesion found in CCE. The widespread acceptance of CCE as a non-invasive diagnostic method is particularly important in the setting of colorectal cancer screening. However, reviewing these exams is a time-consuming process as they generate a large number of frames, with the risk of overlooking important lesions. We aimed to develop an artificial intelligence (AI) algorithm using a convolutional neural network (CNN) architecture for the automatic detection of colonic protruding lesions. A CNN was constructed using an anonymized database of CCE images collected from a total of 124 patients. This database included images of patients with colonic protruding lesions or patients with normal colonic mucosa or with other pathologic findings. A total of 5715 images (2410 protruding lesions, 3305 normal mucosa or other findings) were extracted for CNN development. Two image datasets were created and used for training and validation of the CNN. The performance of the CNN was measured by calculating the area under the receiving operating characteristic curve (AUROC), sensitivity, specificity, positive and negative predictive values (PPV and NPV, respectively). The AUROC for detection of protruding lesions was 0.99. The sensitivity, specificity, PPV and NPV were 90.0%, 99.1%, 98.6% and 93.2%, respectively. The overall accuracy of the network was 95.3%. The developed <jats:italic>deep learning</jats:italic> algorithm accurately detected protruding lesions in CCE images. The introduction of AI technology to CCE may increase its diagnostic accuracy and acceptance for the screening of colorectal neoplasia.</jats:p>

Language: English

Type (Professor's evaluation): Scientific