K2S Challenge: From Undersampled K-Space to Automatic Segmentation

Abstract

Magnetic Resonance Imaging (MRI) offers strong soft tissue contrast but suffers from long acquisition times and requires tedious annotation from radiologists. Traditionally, these challenges have been addressed separately with reconstruction and image analysis algorithms. To see if performance could be improved by treating both as end-to-end, we hosted the K2S challenge, in which challenge participants segmented knee bones and cartilage from 8× undersampled k-space. We curated the 300-patient K2S dataset of multicoil raw k-space and radiologist quality-checked segmentations. 87 teams registered for the challenge and there were 12 submissions, varying in methodologies from serial reconstruction and segmentation to end-to-end networks to another that eschewed a reconstruction algorithm altogether. Four teams produced strong submissions, with the winner having a weighted Dice Similarity Coefficient of 0.910 ± 0.021 across knee bones and cartilage. Interestingly, there was no correlation between reconstruction and segmentation metrics. Further analysis showed the top four submissions were suitable for downstream biomarker analysis, largely preserving cartilage thicknesses and key bone shape features with respect to ground truth. K2S thus showed the value in considering reconstruction and image analysis as end-to-end tasks, as this leaves room for optimization while more realistically reflecting the long-term use case of tools being developed by the MR community.

AI in MRI: Computational Frameworks for a Faster, Optimized, and Automated Imaging Workflow

“…The second trend is the shift towards more comprehensive AI pipelines, which aim to address more than one component of the MRI workflow. It includes frameworks that jointly optimize the sampling pattern and reconstruction [ 45 ] and techniques that generate segmentations directly from undersampled raw MRI measurements, thereby conducting both reconstruction and segmentation [ 76 ].…”

“…Although these two tasks are often addressed separately, there could be much benefit in solving them in tandem. This special issue includes a paper that summarizes the K2S challenge, which focused on this end-to-end approach and was hosted at the 25th International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI) (Singapore, 2022) [ 76 ]. The challenge participants were required to submit DL models that can generate segmentation maps directly from 8x undersampled raw MRI measurements.…”

Synthetic Inflammation Imaging with PatchGAN Deep Learning Networks

“…In all these applications, Gd is administered in standard imaging protocols, so similar datasets can be curated and used to train synthetic post-contrast imaging algorithms to reduce and hopefully eliminate the need for Gd administration. Furthermore, validated algorithms could synthesize post-contrast images from existing large datasets such as the Osteoarthritis Initiative (OAI), K2S, and fastMRI+ to allow for large cohort studies to facilitate a better understanding of inflammation [58][59][60].…”

Joint reconstruction and segmentation in undersampled 3D knee MRI combining shape knowledge and deep learning

“…These bones in fact pose distinct challenges (i) due to significantly lower signal strength and contrast to certain neighboring tissues like tendons as well as (ii) due to their physical size extending outside the field-of-view, thus, covering image regions with higher vulnerability to artefacts like signal decay and nonlinear distortions. Another interesting dataset is the recently published K2S-challange-dataset (Tolpadi et al 2023), which however, has two major limitations: first, ground-truth segmentation was carried out with CNNs rather than manually and a human reader quality rating served as an additional exclusion criterion. In other words, experts identified cases that are convenient for a CNN-based segmentation.…”