POLAR - sporedata/researchdesigneR GitHub Wiki
General description
The POLAR (POLyp Artificial Recognition) database is a dataset designed to assist in the automatic detection and classification of colorectal polyps using artificial intelligence (AI). Colorectal polyps are abnormal growths on the inner lining of the colon or rectum, and their early detection is crucial for preventing colorectal cancer. The POLAR database supports the development of AI models, particularly deep learning and computer vision algorithms, to improve the detection and diagnosis of these polyps during colonoscopies.
The POLAR database aims to enhance AI-based recognition systems for colorectal polyps by providing a large collection of annotated medical images. These systems can support gastroenterologists in improving the accuracy and efficiency of colonoscopies.
Data Categories
The POLAR dataset includes high-quality images and videos captured during colonoscopies. Each image or video segment is labeled with annotations identifying the location, type, and size of polyps. The dataset might also include different types of polyps, such as adenomatous, hyperplastic, and serrated, to support classification algorithms.
Related publications
- Development and validation of a computer-aided diagnosis system for characterization of diminutive colorectal polyps during live endoscopy: a multicentre study with benchmarking against screening endoscopists
- Comprehensive review of publicly available colonoscopic imaging databases for artificial intelligence research: availability, accessibility, and usability
Data access
For more information on the POLAR dataset, visit https://www.amc.nl/web/polar-database.htm