A multi-centre polyp detection and segmentation dataset for generalisability assessment.
Sharib AliDebesh JhaNoha GhatwaryStefano RealdonRenato CannizzaroOsama E SalemDominique LamarqueChristian DaulMichael A RieglerKim Vidar ÅnonsenAndreas PetlundPål HalvorsenJens RittscherThomas de LangeJames Edward EastPublished in: Scientific data (2023)
Polyps in the colon are widely known cancer precursors identified by colonoscopy. Whilst most polyps are benign, the polyp's number, size and surface structure are linked to the risk of colon cancer. Several methods have been developed to automate polyp detection and segmentation. However, the main issue is that they are not tested rigorously on a large multicentre purpose-built dataset, one reason being the lack of a comprehensive public dataset. As a result, the developed methods may not generalise to different population datasets. To this extent, we have curated a dataset from six unique centres incorporating more than 300 patients. The dataset includes both single frame and sequence data with 3762 annotated polyp labels with precise delineation of polyp boundaries verified by six senior gastroenterologists. To our knowledge, this is the most comprehensive detection and pixel-level segmentation dataset (referred to as PolypGen) curated by a team of computational scientists and expert gastroenterologists. The paper provides insight into data construction and annotation strategies, quality assurance, and technical validation.
Keyphrases
- deep learning
- convolutional neural network
- healthcare
- loop mediated isothermal amplification
- end stage renal disease
- real time pcr
- label free
- electronic health record
- ejection fraction
- newly diagnosed
- big data
- palliative care
- peritoneal dialysis
- study protocol
- prognostic factors
- rna seq
- papillary thyroid
- emergency department
- randomized controlled trial
- squamous cell carcinoma
- artificial intelligence
- amino acid
- data analysis
- lymph node metastasis
- clinical evaluation