Endoscopy Disease Detection and Segmentation (EDD2020)¶
Data now online. Please register and download!!!
Google groups: https://groups.google.com/d/forum/endocv2020(please join and help each other!)
-->** Sub-set of test data released. Final set will be released 2 days before the leaderboard closing.**
--> Accepting papers now:
https://cmt3.research.microsoft.com/EndoCV2020 (intention to submit
25th 28th Feb., please note that full test data will be made
available to only these participants)
--> Intention to submit should include, abstract, method brief and your results on current sub-set of dataset (Please do not compare your method with any other fellow participants score in the leaderboard*)
--> *Call for travel grant applications*: *Please send in your application to sharib.ali@eng.ox.ac.uk by 10am, 12th March 2020
FINALS STARTS at 1st March 23:59 till 3rd March 23:59. GOOD LUCK TO ALL PARTICIPANTS.
--> Final full paper submission: 7th March (extended deadline)
ABOUT
Endoscopy is a widely used clinical procedure for the early detection of cancers in hollow-organs such as oesophagus, stomach, colon and bladder. Computer-assisted methods for accurate and temporally consistent localisation and segmentation of diseased region-of-interests enable precise quantification and mapping of lesions from clinical endoscopy videos which is critical for monitoring and surgical planning. Innovations have the potential to improve current medical practices and refine healthcare systems worldwide. However, well-annotated, representative publically-available datasets for disease detection for assessing reproducibility and facilitating standardised comparison of methods is still lacking. Many methods to detect diseased regions in endoscopy have been proposed however these have primarily focussed on the task of polyp detection in the gastrointestinal tract with demonstration on datasets acquired from at most a few data centres and single modality imaging, most commonly white light. Here, we present our multi-class disease detection and segmentation challenge in clinical endoscopy. With this sub-challenge we aim to establish a comprehensive dataset to benchmark algorithms for disease detection.
Specifically we aim to assess: Precise spatio-temporal localisation of disease regions using bounding boxes and exact pixel-level segmentation. Clinical applicability by assessing the online sequential for real-time monitoring and offline performance of algorithms for improved accuracy and better quantitative reporting. Participants will be provided with a large annotated dataset labelled by medical experts and experienced post doctoral researchers with videos and frames from 5 different international centres and 4 organs targeting multiple populations and varied endoscopy video modalities associated with pre-malignant and diseased regions as follows:
Organ 1: Colon, associated disease: polyp, cancer
Organ 2: Oesophagus, associated disease: Barrett’s, dysplasia and
cancer
Organ 3: Stomach, associated disease: pyloric inflammation,
dysplasia
Organ 4: Bladder, associated disease: scars, cancer
Proposed Tasks:
To rigorously assess the problem of disease detection and localisation in various organs on a comprehensive dataset, participants are requested to complete three sub-tasks (single task is not allowed):
1) Disease detection and localisation: This task will be evaluated based on the results of the test dataset provided from a subset of the data collected for training
2) Semantic segmentation: This task will be evaluated based on the results of the test dataset provided from a subset of the data collected for training
3)~~Out-of-sample generalisation: This task will be evaluated based on the results of the test dataset provided exclusive from the training or other test dataset (data from the institution not taken in any other data) ~~
