To develop a deep learning network for extracting features from the blood-supplying regions of the lenticulostriate artery (LSA) and to establish these features as an imaging biomarker for the comprehensive assessment of the lenticulostriate artery-neural complex (LNC). Automatic segmentation of brain regions on T1-weighted images was performed, followed by the development of the ResNet18 framework to extract and visualize deep learning features from three regions of interest (ROIs). The root mean squared error (RMSE) was then used to assess the correlation between these features and fractional anisotropy (FA) values from diffusion tensor imaging (DTI) and cerebral blood flow (CBF) values from arterial spin labeling (ASL). The correlation of these features with LSA root numbers and three disease categories was further validated using fine-tuning classification (Task1 and Task2). Seventy-nine patients were enrolled and classified into three groups. No significant differences were found in the number of LSA roots between the right and left hemispheres, nor in the FA and CBF values of the ROIs. The RMSE loss, relative to the mean FA and CBF values across different ROI inputs, ranged from 0.154 to 0.213%. The model's accuracy in Task1 and Task2 fine-tuning classification reached 100%. Deep learning features extracted from the basal ganglia nuclei effectively reflect cerebrovascular and neurological functions and reveal the damage status of the LSA. This approach holds promise as a novel imaging biomarker for the comprehensive assessment of the LNC.

The risks and prognosis of mild intracerebral hemorrhage (ICH) patients were easily overlooked by clinicians. Our goal was to use machine learning (ML) methods to predict mild ICH patients' neurological deterioration (ND) and 90-day prognosis. This prospective study recruited 257 patients with mild ICH for this study. After exclusions, 148 patients were included in the ND study and 144 patients in the 90-day prognosis study. We trained five ML models using filtered data, including clinical, traditional imaging, and radiomics indicators based on non-contrast computed tomography (NCCT). Additionally, we incorporated the Shapley Additive Explanation (SHAP) method to display key features and visualize the decision-making process of the model for each individual. A total of 21 (14.2%) mild ICH patients developed ND, and 35 (24.3%) mild ICH patients had a 90-day poor prognosis. In the validation set, the support vector machine (SVM) models achieved an AUC of 0.846 (95% confidence intervals (CI), 0.627-1.000) and an F1-score of 0.667 for predicting ND, and an AUC of 0.970 (95% CI, 0.928-1.000), and an F1-score of 0.846 for predicting 90-day prognosis. The SHAP analysis results indicated that several clinical features, the island sign, and the radiomics features of the hematoma were of significant value in predicting ND and 90-day prognosis. The ML models, constructed using clinical, traditional imaging, and radiomics indicators, demonstrated good classification performance in predicting ND and 90-day prognosis in patients with mild ICH, and have the potential to serve as an effective tool in clinical practice. Not applicable.

To develop and validate a machine learning model based on dual-energy computed tomography (DECT) for predicting cervical lymph node metastases (CLNM) in patients diagnosed with nasopharyngeal carcinoma (NPC). This prospective single-center study enrolled patients with NPC and the study assessment included both DECT and 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT). Radiomics features were extracted from each region of interest (ROI) for cervical lymph nodes using arterial and venous phase images at 100 keV and 150 keV, either individually as non-fusion models or combined as fusion models on the DECT images. The performance of the random forest (RF) models, combined with radiomics features, was evaluated by area under the receiver operating characteristic curve (AUC) analysis. DeLong's test was employed to compare model performances, while decision curve analysis (DCA) assessed the clinical utility of the predictive models. Sixty-six patients with NPC were included for analysis, which was divided into a training set (n = 42) and a validation set (n = 22). A total of 13 radiomic models were constructed (4 non-fusion models and 9 fusion models). In the non-fusion models, when the threshold value exceeded 0.4, the venous phase at 100 keV (V100) (AUC, 0.9667; 95 % confidence interval [95 % CI], 0.9363-0.9901) model exhibited a higher net benefit than other non-fusion models. The V100 + V150 fusion model achieved the best performance, with an AUC of 0.9697 (95 % CI, 0.9393-0.9907). DECT-based radiomics effectively diagnosed CLNM in patients with NPC and may potentially be a valuable tool for clinical decision-making. This study improved pre-operative evaluation, treatment strategy selection, and prognostic evaluation for patients with nasopharyngeal carcinoma by combining DECT and radiomics to predict cervical lymph node status prior to treatment.

The differential diagnosis of neurodegenerative dementias is a challenging clinical task, mainly because of the overlap in symptom presentation and the similarity of patterns observed in structural neuroimaging. To improve diagnostic efficiency and accuracy, deep learning-based methods such as Convolutional Neural Networks and Vision Transformers have been proposed for the automatic classification of brain MRIs. However, despite their strong predictive performance, these models find limited clinical utility due to their opaque decision making. In this work, we propose a framework that integrates two core components to enhance diagnostic transparency. First, we introduce a modular pipeline for converting 3D T1-weighted brain MRIs into textual radiology reports. Second, we explore the potential of modern Large Language Models (LLMs) to assist clinicians in the differential diagnosis between Frontotemporal dementia subtypes, Alzheimer's disease, and normal aging based on the generated reports. To bridge the gap between predictive accuracy and explainability, we employ reinforcement learning to incentivize diagnostic reasoning in LLMs. Without requiring supervised reasoning traces or distillation from larger models, our approach enables the emergence of structured diagnostic rationales grounded in neuroimaging findings. Unlike post-hoc explainability methods that retrospectively justify model decisions, our framework generates diagnostic rationales as part of the inference process-producing causally grounded explanations that inform and guide the model's decision-making process. In doing so, our framework matches the diagnostic performance of existing deep learning methods while offering rationales that support its diagnostic conclusions.

Artificial intelligence (AI), particularly deep learning, has demonstrated high diagnostic performance in detecting intracranial aneurysms on computed tomography angiography (CTA) and magnetic resonance angiography (MRA). However, the clinical translation of these technologies remains limited due to methodological limitations and concerns about generalizability. This scoping review comprehensively evaluates 36 studies that applied deep learning to intracranial aneurysm detection on CTA or MRA, focusing on study design, validation strategies, reporting practices, and reference standards. Key findings include inconsistent handling of ruptured and previously treated aneurysms, underreporting of coexisting brain or vascular abnormalities, limited use of external validation, and an almost complete absence of prospective study designs. Only a minority of studies employed diagnostic cohorts that reflect real-world aneurysm prevalence, and few reported all essential performance metrics, such as patient-wise and lesion-wise sensitivity, specificity, and false positives per case. These limitations suggest that current studies remain at the stage of technical validation, with high risks of bias and limited clinical applicability. To facilitate real-world implementation, future research must adopt more rigorous designs, representative and diverse validation cohorts, standardized reporting practices, and greater attention to human-AI interaction.

Functional magnetic resonance imaging (fMRI) has demonstrated significant potential in the early diagnosis and study of pathological mechanisms of Alzheimer's disease (AD). To fit subtle cross-spatiotemporal interactions and learn pathological features from fMRI, we proposed a fine-grained spatiotemporal graph neural network with self-supervised learning (SSL) for diagnosis and biomarker extraction of early AD. First, considering the spatiotemporal interaction of the brain, we designed two masks that leverage the spatial correlation and temporal repeatability of fMRI. Afterwards, temporal gated inception convolution and graph scalable inception convolution were proposed for the spatiotemporal autoencoder to enhance subtle cross-spatiotemporal variation and learn noise-suppressed signals. Furthermore, a spatiotemporal scalable cosine error with high selectivity for signal reconstruction was designed in SSL to guide the autoencoder to fit the fine-grained pathological features in an unsupervised manner. A total of 5,687 samples from four cross-population cohorts were involved. The accuracy of our model was 5.1% higher than the state-of-the-art models, which included four AD diagnostic models, four SSL strategies, and three multivariate time series models. The neuroimaging biomarkers were precisely localized to the abnormal brain regions, and correlated significantly with the cognitive scale and biomarkers (P$< $0.001). Moreover, the AD progression was reflected through the mask reconstruction error of our SSL strategy. The results demonstrate that our model can effectively capture spatiotemporal and pathological features, and providing a novel and relevant framework for the early diagnosis of AD based on fMRI.

MRI sequence classification becomes challenging in multicenter studies due to variability in imaging protocols, leading to unreliable metadata and requiring labor-intensive manual annotation. While numerous automated MRI sequence identification models are available, they frequently encounter the issue of domain shift, which detrimentally impacts their accuracy. This study addresses domain shift, particularly from adult to pediatric MRI data, by evaluating the effectiveness of pre-trained models under these conditions. This retrospective and multicentric study explored the efficiency of a pre-trained convolutional (ResNet) and CNN-Transformer hybrid model (MedViT) to handle domain shift. The study involved training ResNet-18 and MedVit models on an adult MRI dataset and testing them on a pediatric dataset, with expert domain knowledge adjustments applied to account for differences in sequence types. The MedViT model demonstrated superior performance compared to ResNet-18 and benchmark models, achieving an accuracy of 0.893 (95% CI 0.880-0.904). Expert domain knowledge adjustments further improved the MedViT model's accuracy to 0.905 (95% CI 0.893-0.916), showcasing its robustness in handling domain shift. Advanced neural network architectures like MedViT and expert domain knowledge on the target dataset significantly enhance the performance of MRI sequence classification models under domain shift conditions. By combining the strengths of CNNs and transformers, hybrid architectures offer enhanced robustness for reliable automated MRI sequence classification in diverse research and clinical settings. Question Domain shift between adult and pediatric MRI data limits deep learning model accuracy, requiring solutions for reliable sequence classification across diverse patient populations. Findings The MedViT model outperformed ResNet-18 in pediatric imaging; expert domain knowledge adjustment further improved accuracy, demonstrating robustness across diverse datasets. Clinical relevance This study enhances MRI sequence classification by leveraging advanced neural networks and expert domain knowledge to mitigate domain shift, boosting diagnostic precision and efficiency across diverse patient populations in multicenter environments.

'Brain age' is a biological clock typically used to describe brain health with one number, but its relationship with established gradients of cortical organization remains unclear. We address this gap by leveraging a data-driven, region-specific brain age approach in 335 neurologically intact adults, using a convolutional neural network (volBrain) to estimate regional brain ages directly from structural MRI without a predefined set of morphometric properties. Six distinct gradients of brain aging are replicated in two independent cohorts. Spatial patterns of accelerated brain aging in older adults quantitatively align with the archetypal sensorimotor-to-association axis of cortical organization. Other brain aging gradients reflect neurobiological hierarchies such as gene expression and externopyramidization. Participant-level correspondences to brain age gradients are associated with cognitive and sensorimotor performance and explained behavioral variance more effectively than global brain age. These results suggest that regional brain age patterns reflect fundamental principles of cortical organization and behavior.

To investigate whether pulse pressure or mean arterial pressure mediates the relationship between age and white matter hyperintensity load and to examine the mediating effect of white matter hyperintensities on cognition. Demographic information, blood pressure, current medication lists, and Montreal Cognitive Assessment scores for 231 stroke- and dementia-free adults were retrospectively obtained from the Aging Brain Cohort study. Total WMH load was determined from T2-FLAIR magnetic resonance scans using the TrUE-Net deep learning tool for white matter segmentation. In separate models, we used mediation analysis to assess whether pulse pressure or MAP mediates the relationship between age and total white matter hyperintensity load, controlling for cardiovascular confounds. We also assessed whether white matter hyperintensity load mediated the relationship between age and cognitive scores. Pulse pressure, but not mean arterial pressure, significantly mediated the relationship between age and white matter hyperintensity load. White matter hyperintensity load partially mediated the relationship between age and Montreal Cognitive Assessment score. Our results indicate that pulse pressure, but not mean arterial pressure, is mechanistically associated with age-related accumulation of white matter hyperintensities, independent of other cardiovascular risk factors. White matter hyperintensity load was a mediator of cognitive scores across the adult lifespan. Effective management of pulse pressure may be especially important for maintenance of brain health and cognition.

The cerebrospinal fluid (CSF) is known to serve as a unique environment for neurodevelopment, with specific proteins secreted by epithelial cells of the choroid plexus (CP) playing crucial roles in cortical development and cell differentiation. Sex-related differences in the brain in early life have been widely identified, but few studies have investigated the neonatal CSF system and associated transcriptional signatures. This study included 75 full-term neonates [44 males and 31 females; gestational age (GA) = 37-42 weeks] without significant MRI abnormalities from the dHCP (developing Human Connectome Project) database. Deep-learning automated segmentation was used to measure various metrics of the brain ventricular system and CSF. Sex-related differences and relationships with postnatal age were analyzed by linear regression. Correlations between the CP and CSF space metrics were also examined. LASSO regression was further applied to identify the key genes contributing to the sex-related CSF system differences by using regional gene expression data from the Allen Human Brain Atlas. Right lateral ventricles [2.42 ± 0.98 vs. 2.04 ± 0.45 cm3 (mean ± standard deviation), p = 0.036] and right CP (0.16 ± 0.07 vs. 0.13 ± 0.04 cm3, p = 0.024) were larger in males, with a stronger volume correlation (male/female correlation coefficients r: 0.798 vs. 0.649, p < 1 × 10^- 4). No difference was found in total CSF volume, while peripheral CSF (male/female β: 1.218 vs. 1.064) and CP (male/female β: 0.008 vs. 0.005) exhibited relatively faster growth in males. Additionally, the volumes of the lateral ventricular system, third ventricle, peripheral CSF, and total CSF were significantly correlated with their corresponding CP volume (r: 0.362 to 0.799, p < 0.05). DERL2 (Degradation in Endoplasmic Reticulum Protein 2) (r = 0.1319) and MRPL48 (Mitochondrial Large Ribosomal Subunit Protein) (r=-0.0370) were identified as potential key genes associated with sex-related differences in CSF system. Male neonates present larger volumes and faster growth of the right lateral ventricle, likely linked to corresponding CP volume and growth pattern. The downregulation of DERL2 and upregulation of MRPL48 may contribute to these sex-related variations in the CSF system, suggesting a molecular basis for sex-specific brain development.