Magnetic resonance imaging (MRI) offers multiple sequences that provide complementary views of brain anatomy and pathology. However, real-world datasets often exhibit variability in sequence availability due to clinical and logistical constraints. This variability complicates radiological interpretation and limits the generalizability of machine learning models that depend on a consistent multimodal input. Here, we propose an anatomy-guided, modality-agnostic framework to assess disease-related abnormalities in brain MRI, leveraging structural context to ensure robustness in diverse input configurations. Central to our approach is Region ModalMix (RMM), an augmentation strategy that integrates anatomical priors during training to improve model performance under missing or variable modality conditions. Using the BraTS 2020 dataset (n = 369), our framework outperformed state-of-the-art methods, achieving a 9.68 mm average reduction in 95th percentile Hausdorff Distance (HD95) and a 1.36 percentage point improvement in Dice Similarity Coefficient (DSC) over baselines with only one available modality. To evaluate out-of-distribution generalization, we tested RMM on the MU-Glioma-Post dataset (n = 593), which includes heterogeneous post-operative glioma cases. Despite distribution shifts, RMM maintained strong performance, reducing HD95 by 18.24 mm and improving DSC by 9.54% points in the most severe missing-modality scenario. Our framework is applicable to multimodal neuroimaging pipelines, enabling more generalizable abnormality detection under heterogeneous data availability.

Predictive tools for stratifying neonatal hydrocephalus into low- and high-risk groups for cerebrospinal fluid (CSF) diversion are currently lacking. We developed and validated an artificial intelligence (AI) model that integrates multimodal imaging and clinical data to predict CSF diversion needs. The development cohort included 116 neonates with suspicion of raised intracranial pressure (ICP) from a Chinese tertiary referral hospital (80 with intracranial pressure > 80 mm H₂O, 36 with intracranial pressure ≤ 80 mm H₂O). The external validation cohort consisted of 21 neonates with hydrocephalus from an American medical center, categorized by etiology: prenatal myelomeningocele (MMC) closure (n = 5), postnatal MMC closure (n = 6), and post-hemorrhagic hydrocephalus (PHH) (n = 10). Inclusion criteria required available MRI and complete clinical follow-up to confirm CSF diversion outcomes. The primary outcome was the need for CSF diversion. Model performance was assessed using under the receiver operating characteristics curve (AUC), sensitivity, and specificity. The hybrid AI model achieved an AUC of 0.824 in the development cohort in predicting raised ICP, outperforming both the clinical-only model (AUC 0.528, p < 0.001) and the image-only model (AUC 0.685, p = 0.007). In the external validation cohort, the fused MRI-based model achieved an AUC of 0.808. The model correctly predicted CSF diversion in 4/5 prenatal MMC, 4/6 postnatal MMC, and 9/10 PHH cases. The AI model demonstrated robust performance in predicting the need for CSF diversion, particularly in PHH cases, and has the potential to assist decision-making, especially in settings with limited pediatric neurosurgical expertise. Future work should focus on further refining model performance for complex etiologies such as MMC-associated hydrocephalus.

Brain metastases (BMs) are the most common intracranial malignancy, often arising from lung, breast, and melanoma cancers. Receptor tyrosine kinases, such as EGFR and HER2, drive tumor progression and resistance to therapy. Noninvasive detection of these biomarkers, especially in brain metastases, is crucial due to challenges with traditional biopsy methods. This systematic review and meta-analysis assess machine learning (ML)-based models for detecting EGFR mutations and HER2 overexpression in metastatic brain adenocarcinoma using MRI-derived radiomic features. A systematic review and meta-analysis were conducted following PRISMA 2020 guidelines. Studies were identified via PubMed, Scopus, and Web of Science, focusing on ML applications to MRI radiomics for detecting EGFR and HER2 in brain metastases. Data on study design, imaging modality, model type, sample size, and performance metrics were extracted. Subgroup analyses were performed by model type (deep learning vs. classical ML) and sample size (<150 vs. ≥150 participants). A random-effects model was used to pool performance metrics, and risk of bias was assessed using the RoB 2 tool. STATA version 18 and Python 3.10 were used for analyses and visualizations. Of 383 identified studies, 31 (7925 participants) met the inclusion criteria. The pooled analysis showed strong diagnostic performance: AUC = 0.84, accuracy = 0.86, and sensitivity = 0.83. Subgroup analysis revealed higher AUC and accuracy in deep learning models compared with classical ML. Sensitivity analysis also indicated improved AUC in studies with larger sample sizes (≥150), though variability remained. No evidence of heterogeneity or publication bias was detected. ML models demonstrate strong diagnostic performance for detecting EGFR and HER2 in metastatic brain adenocarcinoma, supporting their potential as noninvasive diagnostic tools. However, these findings should be interpreted considering methodological heterogeneity and the limited use of external validation. Further prospective, multicenter studies are warranted to confirm their clinical applicability and generalizability.

Neurodegeneration occurs when the body's central nervous system becomes impaired as a person ages, which can happen at an accelerated pace. Neurodegeneration impairs quality of life, affecting essential functions, including memory and the ability to self-care. Genetics play an important role in neurodegeneration and longevity. Brain age gap estimation (BrainAGE) is a biomarker that quantifies the difference between a machine learning model-predicted biological age of the brain and the true chronological age for healthy subjects; however, a large portion of the variance remains unaccounted for in these models, attributed to individual differences. This study focuses on predicting the BrainAGE more accurately, aided by genetic information associated with neurodegeneration. To achieve this, a BrainAGE model was developed based on MRI measures, and then the associated genes were determined with a Genome-Wide Association Study. Subsequently, genetic information was incorporated into the models. The incorporation of genetic information yielded improvements in the model performances by 7% to 12%, showing that the incorporation of genetic information can notably reduce unexplained variance. This work helps to define new ways of determining persons susceptible to neurological aging decline and reveals genes for targeted precision medicine therapies.

Emerging evidence suggests that blood-oxygen-level-dependent signals in white matter reflect functional activity; however, it remains unclear whether white matter function is altered in rhegmatogenous retinal detachment (RRD) and how it interacts with gray matter. We conducted resting-state functional MRI analyses in patients with RRD and healthy controls to investigate regional white matter activity using amplitude of low-frequency fluctuations/fractional ALFF (ALFF/fALFF), and cross-tissue white matter-gray matter functional connectivity. Voxel-wise analyses were performed to identify aberrant white matter regions, and seed-based connectivity mapping was applied using affected white matter tracts. Support vector machine models were constructed to evaluate the diagnostic utility of these functional features. Patients with RRD exhibited significantly increased ALFF/fALFF in key projection fibers, including the bilateral anterior corona radiata (ACR) and anterior limb of the internal capsule (ALIC). Enhanced functional connectivity was observed between the left ACR and nonvisual gray matter regions such as the right middle temporal gyrus and medial orbitofrontal cortex. Among all features, the fALFF value of the left ALIC demonstrated the highest classification performance (area under the curve = 0.8974) in distinguishing RRD from healthy controls. These findings reveal aberrant spontaneous low-frequency oscillatory activity and enhanced white matter-gray matter coupling in patients with RRD, reflecting cross-tissue functional reorganization beyond the retina. Notably, the elevated fALFF signal in the left ALIC demonstrates strong potential as a neuroimaging biomarker. This study underscores the value of white matter functional metrics in characterizing central nervous system alterations in RRD and offers novel insights into its neurobiological underpinnings.

The aim of the study was to compare the diagnostic quality of deep learning (DL) reconstructed balanced steady-state free precession (bSSFP) single-shot (SSH) cine images with standard, multishot (also: segmented) bSSFP cine (standard cine) in cardiac MRI. This prospective study was performed in a cohort of participants with clinical indication for cardiac MRI. SSH compressed-sensing bSSFP cine and standard multishot cine were acquired with breath-holding and electrocardiogram-gating in short-axis view at 1.5 Tesla. SSH cine images were reconstructed using an industry-developed DL super-resolution algorithm (DL-SSH cine). Two readers evaluated diagnostic quality (endocardial edge definition, blood pool to myocardium contrast and artifact burden) from 1 (nondiagnostic) to 5 (excellent). Functional left ventricular (LV) parameters were assessed in both sequences. Edge rise distance, apparent signal-to-noise ratio (aSNR) and contrast-to-noise ratio were calculated. Statistical analysis for the comparison of DL-SSH cine and standard cine included the Student's t-test, Wilcoxon signed-rank test, Bland-Altman analysis, and Pearson correlation. Forty-five participants (mean age: 50 years ±18; 30 men) were included. Mean total scan time was 65% lower for DL-SSH cine compared to standard cine (92 ± 8 s vs 265 ± 33 s; P  < 0.0001). DL-SSH cine showed high ratings for subjective image quality (eg, contrast: 5 [interquartile range {IQR}, 5-5] vs 5 [IQR, 5-5], P  = 0.01; artifacts: 4.5 [IQR, 4-5] vs 5 [IQR, 4-5], P  = 0.26), with superior values for sharpness parameters (endocardial edge definition: 5 [IQR, 5-5] vs 5 [IQR, 4-5], P  < 0.0001; edge rise distance: 1.9 [IQR, 1.8-2.3] vs 2.5 [IQR, 2.3-2.6], P  < 0.0001) compared to standard cine. No significant differences were found in the comparison of objective metrics between DL-SSH and standard cine (eg, aSNR: 49 [IQR, 38.5-70] vs 52 [IQR, 38-66.5], P  = 0.74). Strong correlation was found between DL-SSH cine and standard cine for the assessment of functional LV parameters (eg, ejection fraction: r = 0.95). Subgroup analysis of participants with arrhythmia or unreliable breath-holding (n = 14/45, 31%) showed better image quality ratings for DL-SSH cine compared to standard cine (eg, artifacts: 4 [IQR, 4-5] vs 4 [IQR, 3-5], P  = 0.04). DL reconstruction of SSH cine sequence in cardiac MRI enabled accelerated acquisition times and noninferior diagnostic quality compared to standard cine imaging, with even superior diagnostic quality in participants with arrhythmia or unreliable breath-holding.

We introduce mpLLM, a prompt-conditioned hierarchical mixture-of-experts (MoE) architecture for visual question answering over multi-parametric 3D brain MRI (mpMRI). mpLLM routes across modality-level and token-level projection experts to fuse multiple interrelated 3D modalities, enabling efficient training without image--report pretraining. To address limited image-text paired supervision, mpLLM integrates a synthetic visual question answering (VQA) protocol that generates medically relevant VQA from segmentation annotations, and we collaborate with medical experts for clinical validation. mpLLM outperforms strong medical VLM baselines by 5.3% on average across multiple mpMRI datasets. Our study features three main contributions: (1) the first clinically validated VQA dataset for 3D brain mpMRI, (2) a novel multimodal LLM that handles multiple interrelated 3D modalities, and (3) strong empirical results that demonstrate the medical utility of our methodology. Ablations highlight the importance of modality-level and token-level experts and prompt-conditioned routing. We have included our source code in the supplementary materials and will release our dataset upon publication.

<i>Objective.</i>Alzheimer's disease (AD) is the most prevalent form of dementia worldwide, encompassing a prodromal stage known as mild cognitive impairment (MCI), where patients may either progress to AD or remain stable. The objective of the work was to capture structural and functional modulations of brain structure and function relying on multimodal MRI data and single nucleotide polymorphisms, also in case of missing views, with the twofold goal of classifying AD patients versus healthy controls and detecting MCI converters.<i>Approach.</i>We propose a multimodal deep learning (DL)-based classification framework where a generative module employing cycle generative adversarial networks was introduced in the latent space for imputing missing data (a common issue of multimodal approaches). Explainable AI method was then used to extract input features' relevance allowing for post-hoc validation and enhancing the interpretability of the learned representations.<i>Main results.</i>Experimental results on two tasks, AD detection and MCI conversion, showed that our framework reached competitive performance in the state-of-the-art with an accuracy of0.926±0.02(CI [0.90, 0.95]) and0.711±0.01(CI [0.70, 0.72]) in the two tasks, respectively. The interpretability analysis revealed gray matter modulations in cortical and subcortical brain areas typically associated with AD. Moreover, impairments in sensory-motor and visual resting state networks along the disease continuum, as well as genetic mutations defining biological processes linked to endocytosis, amyloid-beta, and cholesterol, were identified.<i>Significance.</i>Our integrative and interpretable DL approach shows promising performance for AD detection and MCI prediction while shedding light on important biological insights.

Current measures of adiposity have limitations. Artificial intelligence (AI) models may accurately and efficiently estimate body composition (BC) from routine imaging. To assess the association of AI-derived BC compartments from magnetic resonance imaging (MRI) with cardiometabolic outcomes. Prospective cohort study. UK Biobank (UKB) observational cohort study. 33 432 UKB participants with no history of diabetes, myocardial infarction, or ischemic stroke (mean age, 65.0 years [SD, 7.8]; mean body mass index [BMI], 25.8 kg/m² [SD, 4.2]; 52.8% female) who underwent whole-body MRI. An AI tool was applied to MRI to derive 3-dimensional (3D) BC measures, including subcutaneous adipose tissue (SAT), visceral adipose tissue (VAT), skeletal muscle (SM), and SM fat fraction (SMFF), and then calculate their relative distribution. Sex-stratified associations of these relative compartments with incident diabetes mellitus (DM) and major adverse cardiovascular events (MACE) were assessed using restricted cubic splines. Adipose tissue compartments and SMFF increased and SM decreased with age. After adjustment for age, smoking, and hypertension, greater adiposity and lower SM proportion were associated with higher incidence of DM and MACE after a median follow-up of 4.2 years in sex-stratified analyses; however, after additional adjustment for BMI and waist circumference (WC), only elevated VAT proportions and high SMFF (top fifth percentile in the cohort for each) were associated with increased risk for DM (respective adjusted hazard ratios [aHRs], 2.16 [95% CI, 1.59 to 2.94] and 1.27 [CI, 0.89 to 1.80] in females and 1.84 [CI, 1.48 to 2.27] and 1.84 [CI, 1.43 to 2.37] in males) and MACE (1.37 [CI, 1.00 to 1.88] and 1.72 [CI, 1.23 to 2.41] in females and 1.22 [CI, 0.99 to 1.50] and 1.25 [CI, 0.98 to 1.60] in males). In addition, in males only, those in the bottom fifth percentile of SM proportion had increased risk for DM (aHR for the bottom fifth percentile of the cohort, 1.96 [CI, 1.45 to 2.65]) and MACE (aHR, 1.55 [CI, 1.15 to 2.09]). Results may not be generalizable to non-Whites or people outside the United Kingdom. Artificial intelligence-derived BC proportions were strongly associated with cardiometabolic risk, but after BMI and WC were accounted for, only VAT proportion and SMFF (both sexes) and SM proportion (males only) added prognostic information. None.

Traumatic Brain Injury (TBI) is a global health concern, with mild TBI (mTBI) being the most common form. Despite its prevalence, accurately diagnosing mTBI remains a significant challenge. While advanced neuroimaging techniques like diffusion tensor imaging (DTI) offer promise for more robust diagnosis, their clinical application is limited by inconsistent and heterogeneous post-injury findings. Recently, machine learning (ML) techniques, utilizing DTI metrics as features, have shown increasing utility in mTBI research. This approach helps identify distinct between-group features, paving the way for more precise and efficient diagnostic and prognostic tools. This review aims to analyze studies employing ML techniques to assess changes in DTI metrics after mTBI. Systematic review. We conducted a systematic review, adhering to PRISMA guidelines, on the application of ML with DTI for mTBI diagnosis and prognosis on human subjects. This review identified 36 articles. N/A. Study quality was assessed using the Modified QualSyst Assessment Tool. N/A. The review found ML techniques using DTI Metrics either alone or in combination with other modalities (i.e., structural MRI, functional MRI, clinical scores, or demographics) can effectively classify mTBI patients from controls. These approaches have also demonstrated potential in classifying mTBI patients according to the degree of recovery and symptom severity. In addition, these ML models showed strong predictive power toward cognitive scores and brain structural decline, as quantified by brain-predicted age difference. Larger, externally validated studies are needed to develop robust models for the diagnosis and prognosis of mTBI, using imaging biomarkers (including DTI) in conjunction with non-imaging, on-field, or clinical data. Despite the high predictive performance of ML algorithms, the clinical application remains distant, likely due to the small sample size of studies and lack of external validation, which raises concerns about overfitting. 5. Stage 1.