Deep learning models brain deformations based on atrophy and growth data.
problem Simulating brain deformations due to atrophy and growth.
method Differentiable biomechanical model using deep learning.
result Trained model can rapidly simulate new brain deformations with minimal residuals.
New framework uses EEG to detect brain atrophy in AD, validated on large AD trial.
problem Diagnosis of Alzheimer's disease relies on subjective clinical interpretations.
method Combines Riemannian tangent space mapping and elastic net regression.
result Developed brain atrophy markers validated on large AD trial.
AVRA automatically rates brain atrophy from MRI images, achieving high agreement with human radiologists.
problem Manual visual rating of brain atrophy by radiologists is time-consuming and inconsistent.
method AVRA uses machine learning, trained on 2350 ratings, to automatically rate various atrophy scales.
result AVRA achieves substantial agreement with human radiologists, with Cohen's weighted kappa values of 0.74-0.74.
MAGIC uncovers disease heterogeneity across brain scales.
problem Understanding distinct subtypes of brain diseases at different spatial scales.
method Multi-scale Heterogeneity Analysis and Clustering (MAGIC) using semi-supervised clustering.
result Two main subtypes of AD identified with distinct atrophy patterns.
Novel approach combines local and global brain changes for AD prediction.
problem Detecting Alzheimer's disease through local and global brain changes.
method Patch-based 3D-CNNs combined with global topological features for multi-scale brain tissue connectivity.
result Average precision score of 0.95 for classifying cognitively normal subjects and AD patients (prevalence ~55%).
A brain signature predicts future Alzheimer's dementia in MCI patients.
problem Hard to predict which MCI patients will progress to Alzheimer's dementia.
method Identified a brain signature using machine learning in dementia patients, validated in MCI.
result 90% of MCI individuals with the signature progressed to dementia within 3 years.
DIVE models brain disease progression with high spatial resolution.
problem Reconstruct long-term brain pathology from short-term data.
method Clusters vertex-wise biomarker measurements, estimates average trajectories, and identifies disease-specific patterns.
result Reveals distinct patterns of pathology in different diseases and biomarker types.
Smile-GANs clusters brain MRI scans to reveal disease subtypes and progression.
problem Understanding disease heterogeneity in brain MRI scans.
method Generative Adversarial Networks (GANs) for semi-supervised clustering.
result Discovered four subtypes of Alzheimer's and prodromal phases, with two progressive pathways.
This paper describes a new neuroimaging analysis toolbox that allows for the modeling of nonlinear effects at the voxel level, overcoming limitations of methods based on linear models like the GLM. We illustrate its features using a relevant example in which distinct nonlinear trajectories of Alzheimer's disease relate…
Model learns spatiotemporal patterns on graphs from longitudinal data.
problem Learning spatiotemporal patterns on graphs from longitudinal data.
method Mixed-effects model with stochastic Expectation-Maximization algorithm (MCMC-SAEM).
result Personalized model accurately predicts cortical thickness maps in patients.
Predict and classify brain image evolution trajectories from a single MRI timepoint.
problem Diagnosing early mild cognitive impairment (eMCI) from a single MRI scan.
method Supervised and unsupervised learning frameworks that predict and label intensity patch evolution trajectories from a baseline MRI.
result Classification accuracy increased by up to 10% points compared to single timepoint-based methods.
Study proposes deep learning techniques to diagnose and differentiate Celiac Disease and Environmental Enteropathy from biopsy images.
problem Challenging histopathologic overlap between Celiac Disease and Environmental Enteropathy in biopsy images.
method Color balancing and Random Multimodel Deep Learning (RMDL) architecture to address staining variability.
result Proposed deep learning techniques improve diagnosis accuracy of Celiac Disease and Environmental Enteropathy.
DKT transfers biomarker information between neurodegenerative diseases.
problem Estimating biomarker trajectories in rare neurodegenerative diseases with limited data.
method DKT is a joint-disease generative model that transfers biomarker progressions from common neurodegenerative diseases to rare ones.
result DKT estimates plausible multimodal biomarker trajectories in rare diseases like PCA using only unimodal data.
Method learns shape changes over time from longitudinal data.
problem Tackles learning shape trajectories from repeated observations.
method Combines statistical and deformation models on a diffeomorphism manifold.
result Shows gender and genetic differences in hippocampal atrophy progression.
Deep learning for integrating diverse clinical measurements.
problem Combining data from different measurement instruments in longitudinal clinical registries.
method Domain adaptation using deep learning for mapping items from different instruments.
result Domain adaptation can recover latent trajectories even with limited data and misalignment.
Paper presents a brain tumor segmentation method using NGMM and 3D FVF.
problem Automatic brain tumor segmentation from MRIs is challenging.
method Normalized Gaussian Bayesian classifier and 3D Fluid Vector Flow algorithm.
result The method successfully segments brain tumors from MRI images.
Study investigates deep learning model's reliability in clinical MRI data.
problem Tackles reliability of DL models in clinical out-of-distribution MRI data.
method Investigated performance of DL model trained on diverse datasets compared to clinical data.
result Model performs better in similar protocols but worse in clinical data with different tissue contrasts.
Local semi-supervised method improves brain tissue classification in child MRI.
problem Inaccurate detection of brain tissue classes due to intensity variations in early developing brains.
method Kernel Fisher Discriminant Analysis (KFDA) combined with SSIM for perceptual image quality assessment.
result Optimal brain partitioning into subdomains with different average intensity values and separating surfaces between brain parts.
Study evaluates features and classifiers for brain-computer interface tasks.
problem Improving accuracy in brain-computer interface communication.
method Examined six classical features and twelve classifiers across nine datasets.
result Energy in α and η bands, and Bayesian classifier with Gaussian assumption, outperform other methods.
A graph-based method detects abnormal brain connections in functional MRI data.
problem Detecting abnormal brain connections in functional MRI data.
method High-order Graph Auto-Encoder (GAE) with hypersphere distribution for functional data analysis.
result Identifies correlations between affected brain regions and their simultaneous occurrence over time.
DBGDGM models dynamic brain graphs for better understanding brain function.
problem Previous brain graph models ignore temporal dynamics, limiting their usefulness.
method DBGDGM clusters brain regions into evolving communities and learns dynamic node embeddings.
result DBGDGM outperforms baselines in graph generation, dynamic link prediction, and graph classification.
Brain decoding is a popular multivariate approach for hypothesis testing in neuroimaging. It is well known that the brain maps derived from weights of linear classifiers are hard to interpret because of high correlations between predictors, low signal to noise ratios, and the high dimensionality of neuroimaging data. T…
Extracts causal brain dynamics across multiple scales.
problem Statistical associations do not reflect causal mechanisms in brain dynamics.
method Multiscale causal backbone (MCB) extraction using advanced causal structure learning.
result Sparse MCBs reveal distinct causal roles at different brain frequency bands.
Fast, accurate thalamus segmentation method for MS and ET.
problem Automated thalamic nuclei segmentation for neurological diseases.
method Cascaded multi-planar scheme with modified residual U-Net architecture.
result Statistically significant improvements in thalamus segmentation for MS and ET patients.
A new kernel measures brain network similarities, improving disease classification.
problem Lack of edge weight information in existing graph kernels for brain connectivity networks.
method Ordinal pattern kernel for weighted brain connectivity networks.
result The ordinal pattern kernel achieves better classification performance than state-of-the-art graph kernels.
Study uses machine learning to classify autism based on brain connectivity variability.
problem Classifying autism using brain functional connectivity.
method Machine learning models trained on brain imaging data from ABIDE database.
result Increased FC variability in brain regions associated with low variability in ASD patients.
New method for analyzing brain dynamics using HMMs and graph models.
problem Limited ability of current brain models to explain spontaneous dynamic state changes.
method Hidden Markov Graph Models (HMGMs) and spatiotemporal random walks.
result Identification of important brain community structures.
New method aligns brain data across individuals for better brain decoding.
problem Inter-individual variability in brain response patterns limits decoder generalization.
method SpectralOT method that embeds cortical geometry into Laplace-Beltrami eigenmodes.
result SpectralOT strikes balance between aligning functional features and preserving anatomical structure.
New method predicts AD progression using MEG brain networks.
problem Early diagnosis and prediction of Alzheimer's disease progression.
method MG2G, a deep learning method that maps brain networks into a latent space.
result MG2G detects subtle brain connectivity patterns and predicts AD progression.
Study uses simulation-based inference to decode brain activity from synthetic stimuli.
problem Reversing the process of brain activity emulation to recover stimuli or their properties.
method Pairing brain emulator with LLMs to learn a probabilistic mapping from brain maps to stimulus parameters.
result LLMs can serve as controllable stimulus generators and parameters can be recovered from brain maps.
This research uses Siamese networks to identify partial mouse brain images from the Allen atlas.
problem Identifying precise mouse brain microscopy images from the Allen atlas.
method Siamese Networks with contrastive learning to find corresponding atlas plates for partial images.
result Siamese CNNs achieved 25% TOP-1 and 100% TOP-5 accuracy in identifying brain slices from the Allen atlas.
MarmoNet automates analysis of marmoset brain axonal projections.
problem Automatically detect and segment axonal tracer signals in noisy, cluttered images.
method Uses machine learning, specifically CNNs and image registration, to process and map axonal projections.
result Automated pipeline extracts and maps axonal projections robustly.
A mixture of experts model predicts brain activation from word stimuli.
problem Classical encoding models ignore connections among brain regions.
method Mixture of experts capturing ROI-specific brain activity patterns.
result Model predicts entire brain activation with high spatial accuracy.
Improving the interpretability of brain decoding approaches is of primary interest in many neuroimaging studies. Despite extensive studies of this type, at present, there is no formal definition for interpretability of brain decoding models. As a consequence, there is no quantitative measure for evaluating the interpre…
Study uses deep learning with attention to predict fetal brain gestational age.
problem Accurately predicting fetal brain gestational age for early diagnosis.
method Attention-based deep learning model combining multi-view MRI data.
result Age prediction performance with R2 = 0.94 using multi-view MRI and attention.
Develops a new method to analyze brain networks for cognitive traits.
problem Challenges in summarizing and relating brain connectomes to human traits.
method Graph Auto-Encoding (GATE) model using deep learning.
result GATE improves prediction accuracy and efficiency over existing methods.
Study confirms sparse coding in whole brain using MRI data.
problem Sparse coding in the whole brain's neural activities.
method Applied various matrix factorization methods to fMRI data.
result Sparse coding hypothesis in information representation in the whole human brain is confirmed.
The paper proposes a new model to capture specialized brain regions using mixture of regression experts.
problem Learning a forward mapping that relates stimuli to brain activation assumes all regions respond similarly, ignoring brain specialization.
method Clustering brain regions, learning different linear regression models for each cluster, using a mixture of linear experts.
result The proposed model predicts brain activation more accurately than conventional models.
A new encoding framework predicts brain activity from visual stimuli and intrinsic brain connections.
problem Traditional encoding models ignore brain inner states, limiting their performance in natural image identification.
method Proposes a novel encoding framework combining external stimuli and brain inner states, using a forward encoding model and an inner state model.
result The framework achieves better performance on natural image identification from fMRI responses than traditional models.
Machine learning analysis of neuroimaging data can accurately predict chronological age in healthy people and deviations from healthy brain ageing have been associated with cognitive impairment and disease. Here we sought to further establish the credentials of "brain-predicted age" as a biomarker of individual differe…
CNNs adapted for brain images predict age with improved accuracy.
problem Applying CNNs to brain images without considering their 3D structure.
method Two modifications to existing CNN architectures tailored for brain images.
result Achieved a mean absolute error (MAE) of 1.4 years compared to 1.6 years for a baseline.
Automated brain CT image retrieval from traumatic brain injury cohorts using deep neural networks.
problem Manual image retrieval of whole brain CT scans from large clinical cohorts is time-consuming and resource-intensive.
method Proposes a deep convolutional neural network (dMIR) for automated classification of 2D montage images.
result Achieved high accuracy (f1=1.0) for validation and testing data sets.
Generative model predicts multiple brain graphs from one, preserving topology.
problem Predicting multiple brain graphs from a single one, preserving topology.
method MultiGraphGAN architecture, graph adversarial auto-encoder, cluster-specific decoders, topological loss.
result Significantly outperformed variants in multi-view brain graph generation.
Deep learning extracts brain MRI features without manual intervention.
problem Manual feature extraction from brain MRI data is time-consuming and error-prone.
method Large-scale unsupervised deep learning approach to learn generic feature representations.
result Low-dimensional representations of brain structure with comparable performance to FreeSurfer features.
Framework learns structural and functional brain network embeddings while preserving their properties.
problem Joint learning of structural and functional brain networks while preserving their intrinsic properties.
method Siamese community-preserving graph convolutional network (SCP-GCN) that learns from both structural and functional connectivity.
result Superior performance in neurological disorder analysis compared to existing methods.
New method combines brain imaging data from multiple studies to improve cognitive decoding.
problem Low statistical power in individual neuroimaging studies.
method A new methodology to analyze brain responses across tasks without a unified theoretical framework.
result Improves decoding performance for 80% of 35 functional-imaging studies.
M2E embeds multi-view multi-graph brain networks for better clustering.
problem Clustering brain networks from multiple views to understand disease mechanisms.
method Stack multi-graphs into tensors, use tensor techniques to leverage multi-view and multi-graph interactions.
result M2E outperforms existing methods on clustering brain networks.
DI-SVM improves brain condition decoding performance via domain independence.
problem Transfer learning in brain imaging data with large p and small n.
method DI-SVM minimizes domain dependence via HSIC to learn common features.
result DI-SVM outperforms eight competing methods on brain decoding tasks.