FetReg2021: A Challenge on Placental Vessel Segmentation and Registration in Fetoscopy

  title={FetReg2021: A Challenge on Placental Vessel Segmentation and Registration in Fetoscopy},
  author={Sophia Bano and Alessandro Casella and Francisco Vasconcelos and Abdul Qayyum and Abdesslam Benzinou and Moona Mazher and Fabrice M{\'e}riaudeau and Chiara Lena and Ilaria Anita Cintorrino and Gaia Romana De Paolis and Jessica Biagioli and Daria Grechishnikova and Jing Jiao and Bizhe Bai and Yanyan Qiao and Binod Bhattarai and Rebati Raman Gaire and Ronast Subedi and Eduard Vazquez and Szymon Płotka and Aneta Lisowska and Arkadiusz Sitek and George Attilakos and Ruwan Wimalasundera and Anna Louise David and Dario Paladini and Jan A. Deprest and Elena De Momi and Leonardo S. Mattos and Sara Moccia and Danail Stoyanov},
vessels, tool, fetus and background classes, from 18 in-vivo TTTS fetoscopy procedures and 18 short video clips of an average length of 411 frames for developing placental scene segmentation and frame registration for mosaicking techniques. Seven teams participated in this challenge and their model performance was assessed on an unseen test dataset of 658 pixel-annotated images from 6 fetoscopic procedures and 6 short clips. The challenge provided an opportunity for creating generalized… 
1 Citations

Learning-Based Keypoint Registration for Fetoscopic Mosaicking

The proposed framework achieves higher performance compared to the state of the art, paving the way for robust mosaicking to provide surgeons with context awareness during TTTS surgery.



FetReg: Placental Vessel Segmentation and Registration in Fetoscopy Challenge Dataset

An overview of the Fetreg dataset, challenge tasks, evaluation metrics and baseline methods for both segmentation and registration shows that the dataset poses interesting challenges, offering large opportunity for the creation of novel methods and models through a community effort initiative guided by the FetReg challenge.

Deep Placental Vessel Segmentation for Fetoscopic Mosaicking

A solution utilising the U-Net architecture for performing placental vessel segmentation in fetoscopic videos and a scheme for quantifying drift error in the absence of the ground-truth are proposed.

Deep-learned placental vessel segmentation for intraoperative video enhancement in fetoscopic surgery

A convolutional neural network can be trained to segment placental blood vessels with near-human accuracy and can exceed the accuracy of novice human raters.

Image Registration for Placenta Reconstruction

A method to handle the challenges posed by image registration for placenta reconstruction from fetoscopic video as used in the treatment of Twinto-Twin Transfusion Syndrome (TTTS).

Stable Image Registration for In-Vivo Fetoscopic Panorama Reconstruction

A system can be developed that supports the surgeon during the surgery, by giving feedback and providing a more complete overview of the placenta, by proposing a stable region detection method as well as extracting matchable features based on a deep-learning approach.

Retrieval and registration of long-range overlapping frames for scalable mosaicking of in vivo fetoscopy

This paper proposes the first approach for the construction of mosaics of placenta in in vivo fetoscopy sequences, offering first positive results on in vivo data for which standard mosaicking techniques are not applicable.

Robust fetoscopic mosaicking from deep learned flow fields

This work proposes a fetoscopic mosaicking approach by combining deep learning-based optical flow with robust estimation for filtering inconsistent motions that occurs due to floating particles and specularities, and shows that this approach outperforms both the vessel-based state of the art and LK, noticeably when vessels are either poorly visible or too thin to be reliably identified.

Deep learning-based fetoscopic mosaicking for field-of-view expansion

The proposed mosaicking framework outperformed existing methods and generated meaningful mosaic, while reducing the accumulated drift, even in the presence of visual challenges such as specular highlights, reflection, texture paucity, and low video resolution.

Deep Sequential Mosaicking of Fetoscopic Videos

A new generalized Deep Sequential Mosaicking (DSM) framework for fetoscopic videos captured from different settings such as simulation, phantom, and real environments, which extends an existing deep image-based homography model to sequential data by proposing controlled data augmentation and outlier rejection methods.

Real-time mosaicing of fetoscopic videos using SIFT

Initial qualitative results on ex-vivo placental images indicate that the proposed framework can generate clinically useful mosaics from fetoscopic videos in real-time, and leverages the parallelism of modern GPUs and can process clinical fetoscopic images inreal-time.