FetReg: Placental Vessel Segmentation and Registration in Fetoscopy
Fetoscopy laser photocoagulation is a widely used procedure for the treatment of Twin-to-Twin Transfusion Syndrome (TTTS). The procedure is particularly challenging due to the limited field–of–view, poor manoeuvrability of the fetoscopy, poor visibility due to fluid turbidity and variability in light source, and the unusual position of the placenta. This may lead to increased procedural time and incomplete ablation, resulting in persistent TTTS. Computer-assisted intervention can help overcome these challenges by expanding the fetoscopic field of view through video mosaicking and providing better visualization of the vessel map which in turn can guide the surgeons in better localizing abnormal anastomoses.
To this end, we introduce the first large scale multi-centred dataset of fetoscopy laser photocoagulation procedure, which contains both pixel-wise annotations of fetoscopy frames into three semantic labels and unannotated video clips. The dataset is particularly useful for the semantic segmentation of the fetoscopy environment and for the design and evaluation of endoscopic video mosaicking algorithms with a focus on creating drift-free mosaics. This sub-challenge aims at solving two main sub-tasks:
Task 1: Placental vessel segmentation
Task 2: Placental vessel registration and RGB frame registration for mosaicking
This sub-challenge extends the placental vessel segmentation and mosaicking work presented at MICCAI2020 with the aim of providing benchmark multi-centre data acquired from three major fetal medicine institutions across Europe.
FetReg is
part of the MICCAI2021 EndoVis challenge.
A complete structured description of the accepted MICCAI challenge designs can be found here.
Follow instructions in the Registration section to JOIN and PARTICIPATE in this challenge