We’re thrilled to unveil the iSMIT 2024 Polyp Characterization Challenge, building on the resounding success of previous iterations at ISBI and MICCAI conferences. This year’s challenge, coinciding with the iSMIT conference, offers an exciting opportunity for progress in the field.
Participating teams can choose to tackle one or all of the following tasks: polyp detection in video sequences, high-definition polyp segmentation, and polyp classification. This year we introduce a new dataset, PolypSegm-ASH, featuring an additional histological class—serrated sessile adenomas—aligned with current clinical priorities. This dataset will be instrumental in both polyp segmentation and classification tasks.
Whether you’re pioneering a new methodology or refining your existing approach, the iSMIT 2024 Polyp Characterization Challenge provides an ideal platform for evaluation and innovation.
Accessing the data requires registration on the AI4polypNET website and completion of a data-sharing agreement. Moreover, participation in the challenge and evaluation on the test set require a registration for the iSMIT 2024 conference. We encourage you to submit an abstract showcasing your contributions to medical imaging or related fields and actively participate in this dynamic conference. The AI4polypNET team will be present, offering an excellent opportunity for collaboration and networking. We eagerly await your participation.
Registration opens on Wednesday, May 15th, coinciding with the availability of training data! For all key dates regarding the challenge, please refer to the AI4polypNET website and for all information regarding to iSMIT congress, to its website.
For inquiries or comments, please don’t hesitate to reach out via email at ai4polypnet@cvc.uab.cat. We’re here to support you every step of the way.