Explainable AI for Accelerated Microstructure Imaging: A SHAP-Guided Protocol on the Connectome 2.0 scanner

2025-09-11

AI Summary

Diffusion MRI-based neurite exchange imaging faces a critical bottleneck in clinical and research applications due to excessively long acquisition times (>30 minutes) for gray matter microstructure characterization. To address this, we propose the first interpretable AI-driven protocol optimization framework. Leveraging SHAP values to quantify the information contribution of diffusion-encoding features, we integrate recursive feature elimination with biophysical modeling on the Connectome 2.0 platform to enable rapid, robust estimation of the water exchange time (t_ex). Our data-driven approach identifies an optimal subset of diffusion gradient directions and b-values, reducing scan time to 14 minutes while preserving parameter accuracy—comparable to that of the full protocol—and improving t_ex estimation bias by over twofold without compromising anatomical contrast. This work pioneers the application of interpretable AI to diffusion MRI protocol design, establishing a novel paradigm for efficient, biologically informed microstructural imaging.

Technology Category

• Machine Learning: Optimization
• Search and Optimization: Learning to Search
• Humans and AI: Explainable AI (XAI) for Human Understanding

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Web Mining and Content Analysis: Content-based information diffusion
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

The diffusion MRI Neurite Exchange Imaging model offers a promising framework for probing gray matter microstructure by estimating parameters such as compartment sizes, diffusivities, and inter-compartmental water exchange time. However, existing protocols require long scan times. This study proposes a reduced acquisition scheme for the Connectome 2.0 scanner that preserves model accuracy while substantially shortening scan duration. We developed a data-driven framework using explainable artificial intelligence with a guided recursive feature elimination strategy to identify an optimal 8-feature subset from a 15-feature protocol. The performance of this optimized protocol was validated in vivo and benchmarked against the full acquisition and alternative reduction strategies. Parameter accuracy, preservation of anatomical contrast, and test-retest reproducibility were assessed. The reduced protocol yielded parameter estimates and cortical maps comparable to the full protocol, with low estimation errors in synthetic data and minimal impact on test-retest variability. Compared to theory-driven and heuristic reduction schemes, the optimized protocol demonstrated superior robustness, reducing the deviation in water exchange time estimates by over two-fold. In conclusion, this hybrid optimization framework enables viable imaging of neurite exchange in 14 minutes without loss of parameter fidelity. This approach supports the broader application of exchange-sensitive diffusion magnetic resonance imaging in neuroscience and clinical research, and offers a generalizable method for designing efficient acquisition protocols in biophysical parameter mapping.

Problem

• Reducing long scan times for microstructure imaging
• Optimizing acquisition protocols using explainable AI
• Preserving parameter accuracy with shortened MRI scans

Innovation

• SHAP-guided recursive feature elimination strategy
• Optimized 8-feature subset from 15-feature protocol
• Hybrid framework reduces scan time to 14 minutes