Understanding Benefits and Pitfalls of Current Methods for the Segmentation of Undersampled MRI Data

2025-08-26

AI Summary

Existing undersampled MRI segmentation methods lack a unified benchmark and fair comparative evaluation, leaving the optimal strategy unclear. Method: We establish the first multi-center, multi-dataset benchmark specifically designed for undersampled MRI segmentation, systematically evaluating seven representative approaches—including four one-stage end-to-end models and three two-stage data-consistency methods—using real multi-coil k-space data, standardized preprocessing, training protocols, and evaluation metrics (e.g., Dice score, Hausdorff distance) across two public datasets. Contribution/Results: Our empirical analysis reveals that structurally simple, highly interpretable two-stage methods significantly outperform complex one-stage models. Notably, classical reconstruction techniques (e.g., ESPIRiT) combined with lightweight segmentation networks achieve state-of-the-art performance. This work provides a reproducible benchmark, empirically grounded conclusions, and a practical paradigm for undersampled MRI segmentation.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multi-instance/Multi-view Learning
• Reasoning under Uncertainty: Uncertainty Representations

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

MR imaging is a valuable diagnostic tool allowing to non-invasively visualize patient anatomy and pathology with high soft-tissue contrast. However, MRI acquisition is typically time-consuming, leading to patient discomfort and increased costs to the healthcare system. Recent years have seen substantial research effort into the development of methods that allow for accelerated MRI acquisition while still obtaining a reconstruction that appears similar to the fully-sampled MR image. However, for many applications a perfectly reconstructed MR image may not be necessary, particularly, when the primary goal is a downstream task such as segmentation. This has led to growing interest in methods that aim to perform segmentation directly on accelerated MRI data. Despite recent advances, existing methods have largely been developed in isolation, without direct comparison to one another, often using separate or private datasets, and lacking unified evaluation standards. To date, no high-quality, comprehensive comparison of these methods exists, and the optimal strategy for segmenting accelerated MR data remains unknown. This paper provides the first unified benchmark for the segmentation of undersampled MRI data comparing 7 approaches. A particular focus is placed on comparing extit{one-stage approaches}, that combine reconstruction and segmentation into a unified model, with extit{two-stage approaches}, that utilize established MRI reconstruction methods followed by a segmentation network. We test these methods on two MRI datasets that include multi-coil k-space data as well as a human-annotated segmentation ground-truth. We find that simple two-stage methods that consider data-consistency lead to the best segmentation scores, surpassing complex specialized methods that are developed specifically for this task.

Problem

• Evaluating segmentation methods for undersampled MRI data
• Comparing one-stage versus two-stage segmentation approaches
• Identifying optimal strategies for accelerated MRI segmentation

Innovation

• Benchmarking one-stage and two-stage segmentation approaches
• Using multi-coil k-space data with human-annotated ground-truth
• Finding data-consistency two-stage methods outperform specialized approaches