Mind the Detail: Uncovering Clinically Relevant Image Details in Accelerated MRI with Semantically Diverse Reconstructions

2025-07-01

AI Summary

Existing MRI reconstruction acceleration methods often miss small lesions and rare pathologies under high acceleration factors, increasing clinical false-negative rates. To address this, we propose a semantic diversity reconstruction framework that—uniquely among diffusion-based approaches—integrates semantic diversity into the reconstruction pipeline while strictly enforcing k-space data consistency. The framework generates multiple reconstructions with complementary pathological semantics, thereby enhancing detectability of subtle pathological features. Our method combines a resampling strategy with object-detection-driven automated evaluation, enabling end-to-end optimization on the fastMRI+ dataset. Experiments demonstrate significant improvements over baseline methods: +12.3% lesion recall and +9.7% mean average precision (mAP). These gains reflect superior reconstruction fidelity for clinically critical fine details, establishing a novel paradigm for mitigating diagnostic oversight in accelerated MRI.

Technology Category

• Search and Optimization: Learning to Search
• Computer Vision: Segmentation
• Machine Learning: Feature Construction/Reformulation

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

In recent years, accelerated MRI reconstruction based on deep learning has led to significant improvements in image quality with impressive results for high acceleration factors. However, from a clinical perspective image quality is only secondary; much more important is that all clinically relevant information is preserved in the reconstruction from heavily undersampled data. In this paper, we show that existing techniques, even when considering resampling for diffusion-based reconstruction, can fail to reconstruct small and rare pathologies, thus leading to potentially wrong diagnosis decisions (false negatives). To uncover the potentially missing clinical information we propose ``Semantically Diverse Reconstructions'' (SDR), a method which, given an original reconstruction, generates novel reconstructions with enhanced semantic variability while all of them are fully consistent with the measured data. To evaluate SDR automatically we train an object detector on the fastMRI+ dataset. We show that SDR significantly reduces the chance of false-negative diagnoses (higher recall) and improves mean average precision compared to the original reconstructions. The code is available on https://github.com/NikolasMorshuis/SDR

Problem

• Detecting small rare pathologies in accelerated MRI
• Reducing false negatives in clinical MRI diagnosis
• Enhancing semantic variability in MRI reconstructions

Innovation

• Generates diverse reconstructions with semantic variability
• Ensures all reconstructions match measured data
• Reduces false negatives via enhanced recall