Diffusion Magnetic Resonance Imaging Techniques

Diffusion MRI (dMRI) non-invasively probes the random motion of water molecules to infer tissue microstructure. Fundamental implementations such as diffusion tensor imaging (DTI) assume Gaussian diffusion, yielding quantitative metrics like fractional anisotropy and mean diffusivity that characterise directional coherence and average mobility of water in tissues. Beyond DTI, advanced methods capture non-Gaussian behaviour or resolve complex fibre geometries. Diffusional kurtosis imaging (DKI) extends the tensor model by quantifying the kurtosis of the displacement distribution, offering sensitivity to microstructural heterogeneity. High-angular and multi-shell sampling schemes enable reconstruction of orientation distribution functions, revealing crossing fibres. Techniques such as diffusion spectrum imaging (DSI) trace the full q-space to reconstruct probability density functions of diffusion, thereby visualising intricate fibre configurations. Meanwhile, model-based approaches such as Apparent Measures Using Reduced Acquisitions (AMURA) derive multiple microstructural indices from fewer measurements, balancing acquisition time and parameter richness. Recent innovations in compressed sensing and accelerated acquisition strategies have dramatically shortened scan times without compromising spatial or angular resolution, facilitating translation into clinical and preclinical studies. Collectively, these diffusion MRI techniques have become indispensable tools in neuroscience, neurology and oncology, enabling novel biomarkers for diagnosis, prognosis and treatment monitoring across a broad spectrum of neurological disorders.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Research from all publishers

Recent studies have explored the application of AMURA indices for early detection of Alzheimer’s pathology, demonstrating that reduced-acquisition measures can reliably distinguish amyloid-β and tau deposition groups with supportive explainable artificial intelligence analyses. Another investigation addressed challenges in frequency-dependent diffusional kurtosis imaging by combining an efficient oscillating gradient encoding scheme with axisymmetric modelling and spatial regularisation, enabling robust estimation of directional kurtosis maps from a minimal number of diffusion directions. This innovation offers enhanced sensitivity to microarchitectural alterations in both healthy ageing and disease. Comprehensive reviews of diffusion spectrum imaging have underscored its ability to resolve multidirectional fibre tracts and identify novel neuroimaging biomarkers, while highlighting current limitations in clinical adoption due to lengthy acquisitions and complex post-processing. Proposed solutions include compressed sensing undersampling and extended probability density function modelling to accelerate DSI tractography and improve reproducibility for clinical research.

Diffusion Magnetic Resonance Imaging Techniques publication trend

The graph below shows the total number of articles in diffusion magnetic resonance imaging techniques across all publications each year (not limited to Nature Index journals).

Technical terms

Diffusion tensor imaging (DTI): A model assuming Gaussian water diffusion, yielding tensor-derived metrics of directional water movement.

Diffusional kurtosis imaging (DKI): An extension of DTI that quantifies non-Gaussian diffusion, reflecting tissue heterogeneity.

Diffusion spectrum imaging (DSI): A high-angular resolution method sampling the full q-space to reconstruct diffusion probability density functions.

Apparent Measures Using Reduced Acquisitions (AMURA): A technique extracting multiple diffusion measures from fewer diffusion-weighted images.

Compressed sensing (CS): A reconstruction approach that accelerates image acquisition by exploiting signal sparsity.

b-value: A parameter denoting the strength of diffusion weighting in MRI sequences.

Fractional anisotropy (FA): A scalar index of diffusion directionality within a voxel.

Mean diffusivity (MD): The average diffusion coefficient across all spatial directions in a voxel.

References

1. XAI-Based Assessment of the AMURA Model for Detecting Amyloid-β and Tau Microstructural Signatures in Alzheimer’s Disease. IEEE Journal of Translational Engineering in Health and Medicine (2024).
2. Robust frequency-dependent diffusional kurtosis computation using an efficient direction scheme, axisymmetric modelling, and spatial regularization. Imaging Neuroscience (2024).
3. Research Progress in Diffusion Spectrum Imaging. Brain Sciences (2023).
4. Compressed Sensing Diffusion Spectrum Imaging for Accelerated Diffusion Microstructure MRI in Long-Term Population Imaging. Frontiers in Neuroscience (2018).