2026/05/01 by Maximilian F. Eggl, Silvia De Santis · 2 voices
Medicine · Neuroscience · #Advanced Neuroimaging Techniques and Applications #Functional Brain Connectivity Studies #Glioma Diagnosis and Treatment
paper · pdf · doi:10.1038/s43856-026-01614-6
openalex publication_date 2026/05/01 · openalex created_date 2026/05/02 · openalex updated_date 2026/08/01
Diffusion-weighted magnetic resonance imaging provides a non-invasive way to probe brain tissue microstructure and is widely used in neuroscience and clinical research. Reliable microstructural maps usually require long scan times because many measurements are needed to sample the underlying parameter space. This limits clinical feasibility and accessibility. The aim of this study is to determine whether simulation-based inference can reduce the amount of diffusion data required while preserving estimation fidelity across commonly used diffusion models. We apply simulation-based inference using neural posterior estimation to infer diffusion parameters directly from measured signals. The approach is tested on diffusion tensor imaging, diffusion kurtosis imaging, and biophysical models of axonal density and size. Models are trained entirely on simulated data and evaluated using both simulated datasets and experimental brain data from healthy and pathological individuals. Performance is compared with standard non-linear least squares fitting under noisy and sparsely sampled conditions. Here we show that simulation-based inference achieves reliable parameter estimates using up to 90% fewer measurements than conventional approaches. The method consistently outperforms standard fitting when data are noisy or limited and remains robust across models, sampling schemes, and both healthy and pathological brain data. This study demonstrates that simulation-based inference enables fast and robust microstructural imaging with substantially reduced scan times. The approach supports privacy-preserving workflows, could expand dMRI access, e.g., for pediatric and other time-sensitive patients, enable advanced microstructure-sensitive protocols, and rescue legacy data with suboptimal quality. Magnetic resonance imaging (MRI) is widely used to study the brain, but advanced MRI methods that reveal fine details of brain tissue usually require long scan times. This makes them difficult to use in everyday clinical practice. The aim of this study was to find a way to obtain detailed brain information using much shorter MRI scans. To do this, we used computer simulations and artificial intelligence to learn how to estimate brain tissue properties from limited and noisy MRI data. Instead of training on patient data, the method was trained only on signals designed to imitate those seen in real patient data. We show that reliable brain measurements can be obtained with far fewer MRI scans than usual. This could make advanced MRI faster, more accessible, and easier to use in hospitals. In the future, this approach may help improve diagnosis of disease while reducing scan time and patient burden. Eggl and De Santis develop an AI-based method that estimates detailed brain tissue properties from very short and noisy diZusion MRI scans using only simulated training data. The approach produces accurate results with up to 90% fewer measurements, enabling faster MRI scans and making advanced imaging methods more practical for clinical use.