Motivation: Medical image registration plays a very important role in improving clinical workflows, computer-assisted interventions and diagnosis as well as for research studies involving e.g. morphological analysis. Deep learning for medical registration is currently starting to show promising advances that could improve the robustness, computation speed and accuracy of conventional algorithms to enable better practical translation. Nevertheless, there exists no commonly used benchmark dataset to compare learning based registration among another and with their conventional (not trained) counterparts. With few exceptions (CuRIOUS at MICCAI 2018/2019 and the Continuous Registration Challenge at WBIR 2018) there has also been no comprehensive registration challenge covering different anatomical structures and evaluation metrics. We also believe that the entry barrier for new teams to contribute to this emerging field are higher than e.g. for segmentation, where standardised datasets (e.g. Medical Decathlon) are easily available. In contrast, many registration tasks, require resampling from different voxel spacings, affine pre-registration and can lead to ambiguous and error-prone evaluation of whole deformation fields.
Scope: The Learn2reg challenge, which builds on a popular tutorial in 2019, will be a simplified challenge design that removes many of the common pitfalls for learning and aplying transformations. We will provide pre-preprocessed data (resample, crop, pre-align, etc.) that can be directly employed by most conventional and learning frameworks. Only displacement fields in voxel dimensions in a standard orientation will have to be provided by participants and python code to test their application to training data will be provided as open-source along with all evaluation metrics. Our challenge will consist of 4 clinically relevant sub-tasks (datasets) that are complementary in nature. They can either be individually or comprehensively addressed by participants and cover both intra- and inter-patient alignment, CT, ultrasound and MRI modalities, neuro-, thorax and abdominal anatomies and the four of the imminent challenges of medical image registration:
- learning from small datasets
- estimating large deformations
- dealing with multi-modal scans
- learning from noisy annotations
An important aspect of challenges are comprehensive and fair evaluation criteria. Since, medical image registration is not limited to accurately and robustly transferring anatomical annotations but should also provide plausible deformations, we will incorporate a measure of transformation complexity (the standard deviation of local volume change defined by the Jacobian determinant of the deformation). To encourage the submission of learning based approaches that reduce the computational burden of image registration.. Due to differences in hardware, the computation time (including all steps of the employed pipeline) will be measured by running algorithms on the grand-challenge.org evaluation platform (where the whole challenge will be hosted) with the potential of using Nvidia GPU backends (this will not be a strict requirement for participants).
Organisers / Contact: The full list of organisers can be found in the proposal document below, for any practical questions please contact Adrian Dalca, Alessa Hering, Lasse Hansen and Mattias Heinrich at firstname.lastname@example.org
See the full MICCAI Learn2Reg proposal here