Neuroinformatics最新文献

{"title":"Cross-atlas Identification of Narrative Hubs via Multi-embedding Graph Models in fMRI Data.","authors":"Mohammad Amin Saket, Mansooreh Pakravan","doi":"10.1007/s12021-026-09787-0","DOIUrl":"10.1007/s12021-026-09787-0","url":null,"abstract":"<p><p>One of the main objectives of cognitive neuroscience is to investigate brain processes that underlie narrative comprehension. Furthermore, earlier studies that used naturalistic functional magnetic resonance imaging (fMRI) datasets, like Narratives, has advanced our knowledge of large-scale language and narrative networks, most studies have relied on correlation-based analyses or single-region importance measures, overlooking the dynamic and structural properties of brain networks. In this work, we present a new graph-based framework to identify important regions in narrative comprehension by combining a composite node importance scoring method with multiple node embedding algorithms. We first used controlled simulations with stochastic block models (SBM) with different hub nodes and community strengths to validate the framework. This made it possible to systematically assess seven embedding algorithms for node influence attribution, link prediction, and community detection. Applying the same framework to fMRI data, we analyzed two parcellation schemes, the Harvard-Oxford and Schaefer (100-parcel) atlases, to identify influential cortical regions. Our findings reveal consistent engagement of the default mode, salience, and limbic networks across stories and atlases, emphasizing their central role in narrative processing. Overall, this work offers a reliable, comprehensible method for identifying key brain regions, bridging the gap between graph representation learning and cognitive neuroscience. The framework provides a scalable basis for further research that connects naturalistic cognition, dynamic brain connectivity, and linguistic features.</p>","PeriodicalId":49761,"journal":{"name":"Neuroinformatics","volume":"24 2","pages":""},"PeriodicalIF":2.9,"publicationDate":"2026-05-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147822889","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Diagnostic Classification of Autism Spectrum Disorder in the Frequency Domain Using Resting-State fMRI.","authors":"Hossein Haghighat","doi":"10.1007/s12021-026-09782-5","DOIUrl":"10.1007/s12021-026-09782-5","url":null,"abstract":"<p><p>Autism spectrum disorder (ASD) is a neurodevelopmental disorder with problems in social interactions, verbal and non-verbal communication, repetitive behaviors, and limited interests in a person. Considering the challenges in diagnosing ASD based on behavioral symptoms-such as subjectivity, variability among individuals, and overlap with other developmental conditions-it seems necessary to propose computer-aided diagnosis systems (CADS) for ASD. We proposed an age-dependent CADS based on functional connectivity (FC) in the frequency domain for ASD using resting-state functional magnetic resonance imaging (rs-fMRI). Also, the features and classification accuracy obtained in the frequency and time domains were compared. First, preprocessing was performed on the rs-fMRI data. Then, group-independent component analysis (GICA) was used to obtain resting state networks (RSNs). This was followed by obtaining separate components of RSNs for each individual using dual regression. Then, coherence analysis was used to extract the features of FC in the frequency domain between RSNs. To consider the role of age in the classification process, three age groups of children, adolescents, and adults were considered, and feature selection for each age group was applied separately using an embedded approach, in which all classifiers in the Waikato Environment for Knowledge Analysis (WEKA) machine learning platform were used simultaneously. Finally, classification accuracy was obtained for each age group. The proposed CADS was able to classify 95.23% in the children group, 88.1% in the adolescent group, and 92.8% in the adult group. In addition, the frequency bands whose features obtained the most distinction in each age group were identified, highlighting their potential relevance for supporting ASD diagnosis and monitoring rehabilitation.</p>","PeriodicalId":49761,"journal":{"name":"Neuroinformatics","volume":"24 2","pages":""},"PeriodicalIF":2.9,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147822855","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Template-Based Label Propagation for Mouse Brain MRI Skull Stripping.","authors":"Rui Gong, Andrii Gegliuk, Daria Sharapova, Bhanu Sharma, Yoshifumi Abe, Ken Nakae, Shin Ishii, Sho Yagishita","doi":"10.1007/s12021-026-09785-2","DOIUrl":"10.1007/s12021-026-09785-2","url":null,"abstract":"<p><p>Accurate skull stripping is an essential preprocessing step in mouse brain magnetic resonance imaging, particularly for reliable atlas registration and large scale population studies. Existing approaches are often labor intensive, sensitive to inter subject variability, and typically require manual brain masks for many individual subjects. We present a high throughput skull stripping pipeline that utilize template-based label propagation to efficiently generate training data for automated segmentation. A population average ex vivo T2-weighted mouse brain MRI template including the skull was constructed, and a single brain mask was manually annotated in template space. This mask was propagated to individual subjects using inverse transformations and used to train an attention-based 3D U-Net segmentation model. Compared with conventional pipelines requiring manual masks from multiple subjects, the proposed approach achieves competitive segmentation performance while substantially reducing manual annotation effort. Additional experiments comparing training with propagated labels, manual labels, and their combination showed that training on propagated labels alone provided robust performance, suggesting that label consistency may be as important as label quality. To evaluate generalizability, we conducted experiments on an independent in vivo mouse MRI dataset. Direct application of the trained model using ex vivo data to in vivo data resulted in reduced performance, indicating a domain shift between imaging conditions. However, applying the full pipeline to the in vivo dataset, including template construction and label propagation, yielded strong segmentation performance. These results indicate that, while trained models are domain specific, the proposed framework is adaptable across imaging conditions and provides a practical strategy for generating large, anatomically consistent training datasets for biomedical image segmentation.</p>","PeriodicalId":49761,"journal":{"name":"Neuroinformatics","volume":"24 2","pages":""},"PeriodicalIF":2.9,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13135007/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147822869","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unveiling an ALS Blood Transcriptomic Signature: A Machine Learning Classifier Distinct from Neurodegenerative Controls.","authors":"Elisa Gascón, Ana Cristina Calvo, Pilar Zaragoza, Rosario Osta","doi":"10.1007/s12021-026-09780-7","DOIUrl":"10.1007/s12021-026-09780-7","url":null,"abstract":"<p><p>The absence of accessible and reliable biomarkers constitutes a critical barrier for the early diagnosis and stratification of neurodegenerative diseases. While peripheral blood offers a minimally invasive window into systemic pathophysiology, identifying molecular signatures that survive biological heterogeneity and technical noise remains an unresolved challenge. In this study, this issue was addressed through a comparative systemic transcriptomic analysis of Amyotrophic Lateral Sclerosis (ALS), Alzheimer’s disease (AD), and Parkinson’s disease (PD) in whole blood, implementing a comprehensive workflow integrating unsupervised network analysis and supervised machine-learning methods. By employing LASSO regression and cross-validation across independent external cohorts, a stable and specific transcriptomic signature for ALS was identified, comprising key crosstalk genes involved in systemic immune dysregulation and microglial function, including CTSS, PTEN, IL18, PTPRC, and CSF1R. In contrast, AD and PD exhibited weak transcriptomic signatures with poor predictive reproducibility, suggesting a distinctive systemic pathology in ALS. In addition, the study confirms the superiority of linear modeling for this genomic signature: while complex non-linear algorithms, specifically Radial Basis Function (RBF) kernel Support Vector Machine (SVM) and Random Forest, displayed high initial performance, they collapsed due to overfitting during external validation. Conversely, the linear LASSO model demonstrated superior robustness and generalizability (AUC 0.74). In conclusion, this study not only defines a unique systemic immunotranscriptomic signature for ALS, distinguishable from other neurodegenerative pathologies, but also establishes interpretability and linear simplicity as essential factors for developing reproducible blood-based biomarkers with clinical translational potential.</p>","PeriodicalId":49761,"journal":{"name":"Neuroinformatics","volume":"24 2","pages":""},"PeriodicalIF":2.9,"publicationDate":"2026-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13124960/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147787447","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cross-Platform Neurotransmitter & Alias Ambiguity for OA-AL2b1 and OA-AL2b2 Neurons in Drosophila melanogaster.","authors":"Renoy Antony Olivera","doi":"10.1007/s12021-026-09783-4","DOIUrl":"10.1007/s12021-026-09783-4","url":null,"abstract":"<p><p>Accurate neuron identification is necessary for reproducible connectomics. While examining historically described octopaminergic neurons of the Drosophila melanogaster optic lobe, I found that most octopaminergic (OA) neurons of interest remained searchable under their original names across commonly used resources. However, two neurons did not behave the same way. OA- AL2b1, which was linked to the alias LoVCLo3, and OA-AL2b2, which was linked to the alias MeVCMe1. This created a selective nomenclature problem. OA-AL2b1 was especially notable because it remained octopaminergic in the queried resources, yet its historical OA-based name was not consistently preserved as the searchable or visible label. OA-AL2b2 showed a different pattern; in current connectomic tools, it was labeled cholinergic, whereas in other databases, it still remained under octopaminergic groupings. Importantly, Busch et al. originally noted that OA-AL2b2 was not confirmed to be octopamine immunoreactive. An additional layer of ambiguity arose because neuPrint displayed a predicted neurotransmitter (NT) field, whereas Neuroglancer displayed consensus NT, both showing Acetylcholine as its NT. Together, these observations show how small inconsistencies in nomenclature and annotation can create major practical problems in neuron retrieval, interpretation, and cross-platform reproducibility. Although this report focuses on two neurons in Drosophila, the same problem can arise more broadly whenever historical names, database aliases, and current annotation systems are not interlinked.</p>","PeriodicalId":49761,"journal":{"name":"Neuroinformatics","volume":"24 2","pages":""},"PeriodicalIF":2.9,"publicationDate":"2026-04-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147787450","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Quantifying Correlogram Shape to Analyze Neuronal Firing Dynamics Recorded in TBI-on-a-Chip.","authors":"Casey Erin Adam, Shatha J Mufti, Jhon Martinez, Edmond A Rogers, Martina Dalolio, Nikita Krishnan, Timothy Beauclair, Riyi Shi","doi":"10.1007/s12021-026-09770-9","DOIUrl":"10.1007/s12021-026-09770-9","url":null,"abstract":"<p><p>Changes in neuronal network dynamics in response to different treatments or conditions underly brain function and pathology. A multitude of tools, including electroencephalography, voltage or calcium imaging, and microelectrode array (MEA) recordings, exist to record signals from neuronal firing at different length scales. Correlograms are a standard analysis tool to study the relationship between a pair of recorded neuronal signals. Correlogram shape can provide information about signal independence, firing pattern, and firing order. However, such analysis is performed manually and qualitatively, limiting the amount of information gained. To overcome this limitation, a MATLAB algorithm was developed to automate correlogram shape quantification by calculating correlogram uniformity, peak count and location, and area left of zero, which respectively quantify signal independence/dependence, firing pattern, and firing order. Algorithm outputs were validated using three different MEA recordings during which cells were exposed to bicuculline methiodide, pH shock, or impact injury. Algorithm outputs described signaling changes in all three recordings, bridged changes in individual signal pairings to changes in the entire signal population, and agreed with literature studies. Therefore, this algorithm serves as a useful means of automatically quantifying changes in signal dependence, firing pattern, and firing order across time within a single recording, and across different recordings. These features are also common to all recording techniques, and therefore can be compared across different types of recordings. Therefore, the MATLAB algorithm described in this article can help provide insight into how neural network dynamics are altered by different drugs or conditions.</p>","PeriodicalId":49761,"journal":{"name":"Neuroinformatics","volume":"24 2","pages":""},"PeriodicalIF":2.9,"publicationDate":"2026-04-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13111530/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147787438","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Corpus Callosum Segmentation on Structural MRI Using Multi-atlas Deformable Registration.","authors":"Amir Sariaslani, Hamid Abrishami Moghaddam, Babak A Ardekani","doi":"10.1007/s12021-026-09778-1","DOIUrl":"10.1007/s12021-026-09778-1","url":null,"abstract":"<p><p>The corpus callosum (CC), the largest white matter structure in the brain, exhibits morphological changes in normal aging and neurodegenerative disorders. A reliable CC segmentation method facilitates early detection and monitoring of neurological disorders such as Alzheimer’s disease (AD). This study presents Yuki, a fully automated CC segmentation program, evaluates its performance, and demonstrates its utility in structural MRI analysis. We describe an automated method for CC segmentation on structural MRI using multi-atlas deformable registration. Accuracy is assessed in 90 participants using the Dice Similarity Coefficient (DSC). Cross-sectionally, 100 cognitively normal (CN) and 100 AD participants are compared in CC area, circularity, and thickness. Longitudinally, test-retest reliability is evaluated in 49 participants with back-to-back scans, and CC circularity decline is analyzed over two years in 62 CN and 55 AD participants. Yuki achieved a DSC of 97.4% on the test set and 98.2% in the test-retest analysis. Cross-sectionally, CC area, circularity, and thickness were significantly reduced in AD compared to CN (p < .001, p < .0001, p < 10⁻6). Point-wise analysis showed diffuse CC thinning in AD except at the isthmus. CC circularity declined significantly over two years in both CN and AD, with a faster rate in AD (p < 10⁻15, p < 10⁻15, p < 10⁻4). Yuki provides accurate, reliable, and reproducible CC segmentation for 3D MRI. Its application in AD highlights its utility in neurodegenerative research, where declines in CC thickness and circularity can serve as structural biomarkers of interest.</p>","PeriodicalId":49761,"journal":{"name":"Neuroinformatics","volume":"24 2","pages":""},"PeriodicalIF":2.9,"publicationDate":"2026-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147787413","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluating Homoscedastic Uncertainty Weighting and Generative Tau Imputation in a Multimodal, Multitask Deep Learning Framework for Alzheimer's Disease.","authors":"Romana Caposova, Tim Blackwell","doi":"10.1007/s12021-026-09781-6","DOIUrl":"10.1007/s12021-026-09781-6","url":null,"abstract":"<p><p>Alzheimer’s disease is a progressive neurodegenerative disorder and the leading cause of dementia. Multimodal longitudinal deep learning models are increasingly used to forecast disease stage and clinical decline, yet it remains unclear whether popular methodological refinements meaningfully improve performance in practice. We analyzed data from 2,271 participants in the Alzheimer’s Disease Neuroimaging Initiative. A multimodal CNN–BiLSTM model was trained jointly to predict diagnostic stage and four clinical outcomes: Alzheimer’s Disease Assessment Scale, Mini-Mental State Examination, Functional Activities Questionnaire, and Clinical Dementia Rating–Sum of Boxes. We evaluated three commonly used strategies: homoscedastic uncertainty weighting for multitask learning, saliency-guided snapshot analysis, and conditional variational autoencoder–mixture density network imputation of missing longitudinal tau trajectories. Preprocessing was implemented in a split-safe manner, with all preprocessing objects fit on training data only and then applied unchanged to validation and test data. Uncertainty weighting showed a modest descriptive classification benefit, particularly in the no-tau setting, but did not provide a consistent regression advantage across endpoints. Learned task weights were stable across splits and assigned lower relative precision to classification than to the regression heads. Saliency and occlusion analyses indicated that cognitive and magnetic resonance imaging inputs carried the largest aggregated attribution, whereas the tau branch contributed negligibly when included. Snapshot models retained some classification signal but underperformed the intact model and did not preserve regression performance. The tau imputation model reconstructed sparse tau trajectories reasonably well, but adding imputed tau did not improve downstream prediction. In this CNN–BiLSTM pipeline, added methodological complexity yielded mixed and limited practical benefit, underscoring the importance of empirically testing refinements rather than assuming they improve multimodal Alzheimer’s disease forecasting.</p>","PeriodicalId":49761,"journal":{"name":"Neuroinformatics","volume":"24 2","pages":""},"PeriodicalIF":2.9,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147693155","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Neural Analysis Toolkit Unifies Semi-Analytical Techniques to Simplify, Understand, and Simulate Dendrites.","authors":"Willem A M Wybo","doi":"10.1007/s12021-025-09766-x","DOIUrl":"10.1007/s12021-025-09766-x","url":null,"abstract":"<p><p>While simulating compartmental dynamics in response to various input patterns is the prevalent technique for understanding dendritic computation, a great deal can be learned from classical analytical methods that provide solutions for the dendritic voltage. For example, such solutions are needed to simplify spatially extended neuron models, to understand frequency-dependent response properties, to elucidate the interaction between synaptic inputs, and hence to reveal the effective compartmentalization of dendrites into functional subunits. Nevertheless, these methods have not been implemented in modern software tools. This work describes the NEural Analysis Toolkit (NEAT), a Python toolbox that implements classical algorithms to compute response properties of spatially extended neuron models, and that leverages these algorithms to simplify them. Packaged with this are a range of useful utilities to plot morphologies and spatial quantities defined on the morphology, to distribute locations on the morphology, and to select parts of the morphology to e.g. apply morphological ablations or alter the membrane properties. The resulting models can be exported to NEURON and NEST, two commonly used simulators, the former focused on detailed single neuron model simulations, and the latter geared towards distributed network simulations. As a consequence, this toolbox provides a missing link between single neuron computation and large-scale network analysis, substantially facilitating the study of the role of dendritic computation in shaping emergent network dynamics.</p>","PeriodicalId":49761,"journal":{"name":"Neuroinformatics","volume":"24 2","pages":""},"PeriodicalIF":2.9,"publicationDate":"2026-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13079553/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147678415","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Full Model Optimisation of the Processing Pipeline in Functional Near-Infrared Spectroscopy.","authors":"Robert J Ward, Gustavo Rodriguez-Gomez, Felipe Orihuela-Espina","doi":"10.1007/s12021-026-09772-7","DOIUrl":"10.1007/s12021-026-09772-7","url":null,"abstract":"<p><p>The analysis of functional Near-Infrared Spectroscopy (fNIRS) data relies on complex processing pipelines composed of multiple functions and parameters, yet the design of such pipelines remains ad hoc and heterogeneous across the field. This lack of standardisation undermines reproducibility and hinders objective comparison of findings. Here, we introduce a general optimisation framework for full model selection (FMS) of fNIRS pipelines, formulated as a hybrid discrete–dense optimisation problem. Our approach leverages gradient-based methods to jointly optimise function choices and parameter values, enabling systematic exploration of the mixed search space without reducing it to purely discrete approximations. To alleviate combinatorial explosion in the discrete search space, we introduce a graph pruning step that reduces the number of valid node connections without excluding admissible solutions. We validate the framework using synthetic and proof-of-concept NIRS datasets, demonstrating its ability to recover plausible pipelines and its flexibility with respect to user-defined criteria such as accuracy, physiological plausibility, or reproducibility. While illustrated in fNIRS, the framework is broadly applicable to neuroimaging and other domains where pipelines integrate categorical and continuous components. By shifting pipeline design from artisanal practice to principled optimisation, this work contributes to the development of reproducible and standardised neuroimaging workflows.</p>","PeriodicalId":49761,"journal":{"name":"Neuroinformatics","volume":"24 2","pages":""},"PeriodicalIF":2.9,"publicationDate":"2026-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147678382","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}