Minimally invasive surgery using cameras to observe the internal anatomy is the preferred approach to many surgical procedures. As a result, endoscopic image processing and surgical vision are evolving as techniques needed to facilitate computer assisted interventions (CAI). Algorithms that have been reported for endoscopic images and video include 3D surface reconstruction, salient feature motion tracking instrument detection or surgical workflow recognition. However, what is missing so far are common datasets for consistent evaluation and benchmarking of algorithms against each other. In computer vision outside surgical applications, such strategies are common place but this has not yet been achieved for endoscopic data. As an endoscopic vision CAI challenge at MICCAI, our aim is to provide a formal framework for evaluating the current state-of-the-art, gather researchers in the field and provide high quality data with protocols for validating endoscopic vision algorithms.
We invite the community to be part of organizing this challenge by contributing data for a specific sub-challenge in the field of endoscopic image processing and surgical vision (e.g. 3D surface reconstruction, tissue classification, feature or instrument tracking). All data used in the challenge will be made publically available and should comply with the quality guidelines described here. The provider of the sub-challenge data will coordinate a corresponding journal paper summarizing the results, contributions and the data itself.
If you are interested in contributing your data and proposing a sub-challenge, please fill out the checklist and send it to endoscopy-challenge@iar.kit.edu by March 24th 2017. Submitted forms will be reviewed by the challenge chairs. Notification of acceptance will be provided by March 30th. We are looking forward to hearing from you!
Stefanie Speidel, Lena Maier-Hein and Danail Stoyanov (Challenge chairs)
Leave a Reply