CAHAL: Clinically Applicable resolution enHAncement for Low-resolution MRI scans

2026-04-20

AI Summary

Clinical routine MRI scans are typically acquired with thick, anisotropic slices, which limits the accuracy of brain morphometric analysis. Existing super-resolution methods often introduce anatomical hallucinations and structural distortions. To address these challenges, this work proposes an anti-hallucination, physics-informed resolution enhancement framework that performs reconstruction directly in the original acquisition space. The approach innovatively integrates a bivariate mixture-of-experts architecture—conditioned on both resolution and anisotropy—with a deterministic routing mechanism and a multi-objective constrained loss. It combines residual 3D U-Net experts, edge-penalized reconstruction, Fourier spectral consistency, and segmentation-guided semantic constraints, trained using a physical degradation model derived from real clinical data. Evaluated on T1 and FLAIR sequences, the method outperforms current generative baselines in both reconstruction fidelity and computational efficiency, significantly enhancing the clinical applicability of isotropic reconstructions.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multimodal Learning
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Large-scale automated morphometric analysis of brain MRI is limited by the thick-slice, anisotropic acquisitions prevalent in routine clinical practice. Existing generative super-resolution (SR) methods produce visually compelling isotropic volumes but often introduce anatomical hallucinations, systematic volumetric overestimation, and structural distortions that compromise downstream quantitative analysis and diagnostic safety. To address this, we propose CAHAL (Clinically Applicable resolution enHAncement for Low-resolution MRI scans), a hallucination-robust, physics-informed resolution enhancement framework that operates directly in the patient's native acquisition space. CAHAL employs a deterministic bivariate Mixture of Experts (MoE) architecture routing each input through specialised residual 3D U-Net experts conditioned on both volumetric resolution and acquisition anisotropy, two independent descriptors of clinical MRI acquisition. Experts are optimised with a composite loss combining edge-penalised spatial reconstruction, Fourier-domain spectral coherence matching, and a segmentation-guided semantic consistency constraint. Training pairs are generated on-the-fly via physics-based degradation sampled from a large-scale real-world database, ensuring robust generalisation. Validated on T1-weighted and FLAIR sequences against generative baselines, CAHAL achieves state-of-the-art results, improving the best related methods in terms of accuracy and efficiency.

Problem

• super-resolution
• MRI
• anisotropic acquisition
• anatomical hallucination
• quantitative analysis

Innovation

• super-resolution
• Mixture of Experts
• physics-informed
• anisotropic MRI
• hallucination-robust