Neuroinformatics最新文献

The open field test is widely used in behavioral neuroscience, providing insights into exploration, anxiety, and the learning processes associated with habituation to novelty. Analyses of exploratory behaviors in open field areas rely heavily on movement changes over time. These activity measures are susceptible to confounds from group differences in locomotor abilities and only provide an indirect measure of learning during exploration. Considerable effort has been placed on identifying additional measures of behavior that can better describe changes in exploration and habituation of novelty. Two measures for enhanced analysis of exploration are coverage and directional persistence (P₊₊). Coverage measures the number of visits to segments of the arena boundary and represents the number of opportunities to habituate to the novelty of this boundary. P₊₊ measures the probability of continued movement in the same direction, reflecting goal-directed exploration, which decreases as the animal habituates the novel arena. Our new Python package, opynfield, calculates coverage, P₊₊, and activity measures from open field tracking data. We further introduce versions of coverage and the analysis of additional motion probabilities. The package includes new, in-depth statistical approaches and data visualizations. We demonstrate the applicability of opynfield using experiments with Drosophila melanogaster in which we (1) validate opynfield's statistical tests, (2) substantiate coverage as a measure of novelty habituation, and (3) characterize behavioral differences in exploration. We also illustrate the utility of opynfield for analyzing rodent exploration by applying it to data from an experiment with Mus musculus. By leveraging full-density tracking data, opynfield facilitates a more nuanced understanding of exploration, potentially leading to improved insights into animal behavior and changes in learning, locomotor activity, and anxiety.

This computational study aimed to optimize the theta burst stimulation (TBS) protocols by systematically exploring the effects of novel frequency couplings combining alpha-band bursts (10 Hz) with pulses in beta (21-29 Hz) and gamma (30-100 Hz) ranges on cortical excitability. Utilizing a revised calcium-dependent plasticity model, we simulated intermittent (iTBS) and continuous (cTBS) TBS after-effects under conventional (5 Hz burst, 50 Hz pulse), Nyffeler's modified (6 Hz burst, 30 Hz pulse), and proposed alpha-beta/gamma frequency couplings. Model robustness was assessed via sensitivity analyses. Novel alpha-beta/gamma couplings consistently induced more pronounced Motor-Evoked Potential (MEP) after-effects. For iTBS/cTBS, alpha-beta coupling (10 Hz burst, 21 Hz pulse) yielded the highest facilitatory/inhibitory effect (14.25/-93.17), markedly surpassing Nyffeler's (7.71/-8.81) and conventional (5.48/-5.35). Alpha-gamma coupling (10 Hz burst, 30 Hz pulse) also showed superior effects. Sensitivity and uncertainty analyses confirmed higher responsiveness. Coupling alpha-band bursts with targeted beta/gamma pulse frequencies markedly enhances the efficacy of TBS-induced cortical plasticity. These findings provide a strong computational rationale for empirical validation and potential clinical translation to improve neuromodulation precision in neuropsychiatric disorders. This work introduces promising optimized TBS protocols that may elevate therapeutic outcomes and reduce treatment variability, advancing non-invasive brain stimulation interventions.

Modelling real-world networks allows investigating the structure and the dynamics of such networks, which led to significant developments in various scientific fields. One of the most used models in these investigations is the Watts-Strogatz, with a structure composed of high clustering and short path lengths known as small-world networks. This model proposes an interesting gradient between regular and random networks, but its generating process, which relies on a single rewiring probability parameter, is hard to access and to manipulate. In order to study the mechanics of the Watts-Strogatz model, the present work proposes a new method based on deep neural networks that could estimate its probability p. To illustrate its applicability, neuroimaging and phenotypic resting-state fMRI data were used from patients with ADHD and typical development children, obtained from the ADHD-200 database. The neural network efficiently estimated the probability parameter, resulting in small-world graphs for functional brain connectivity with a mean ± s.e.m. p distribution of 0.804 ± 0.003. Despite no difference was found considering the gender or diagnosis of participants, the generalized linear model revealed age as a significant predictor of p (mean ± s.e.m.: 4.410 ± 0.877; p < 0.001), indicating a great effect of neurodevelopment on the brain network's structure. The proposed approach is promising in estimating the probability of the Watts-Strogatz model, and its application has the potential to improve investigations of network connectivity with a relatively efficient and simple framework.

The fasciola cinerea (FC) is a slender archicortical band at the posterior hippocampal tail, and its human morphology and network role are poorly defined. To generate a reproducible in vivo three-dimensional (3D) model of the FC, quantify its geometry, characterize structural and functional connectivity within posterior-medial memory networks, and test a tractography-constrained computational model in which the FC acts as a multiplicative gate. Open 7 T datasets, structural, diffusion, and resting-state functional magnetic resonance imaging (fMRI) were anchored to BigBrain and Julich-Brain priors. A semi-automated, atlas-guided pipeline was used to segment the FC and derive morphometrics (volume, thickness, width, curvature, and Laplace-Beltrami spectral shape). Reliability was assessed using the Dice, 95% Hausdorff distance, and test-retest intraclass correlation coefficient (ICC). Diffusion tractography was used to estimate the FC structural pathways toward retrosplenial (RSC), parahippocampal (PHC), posterior cingulate (PCC), and thalamic targets. Resting-state coupling was summarized using Fisher-z correlations and narrowband coherence. A Wilson-Cowan neural mass model, constrained by tractography, simulated FC-dependent FC-RSC coherence with morphometric scaling of gating gain. Segmentation was reliable (Dice = 0.78 ± 0.05; 95% Hausdorff = 1.62 ± 0.41 mm; ICC_volume = 0.88; ICC_thickness = 0.82). Group morphometrics: volume 84.3 ± 17.9 mm³, mean thickness 0.92 ± 0.15 mm, width 1.86 ± 0.31 mm, centerline length 14.2 ± 2.1 mm. FC showed preferential connectivity: FC→RSC 0.21 ± 0.09; FC→PHC 0.18 ± 0.08; FC→PCC 0.11 ± 0.06; FC→Thalamus 0.06 ± 0.04. Resting-state coupling was strongest for FC-RSC (z = 0.24 ± 0.12) with a slow-band coherence enhancement. Thickness predicted the FC→RSC strength (β = 0.17 per 0.1 mm) and FC-RSC z (β = 0.08 per 0.1 mm), and higher curvature was negatively related. The gating model reproduced empirical FC-RSC coherence (r = 0.52 ± 0.11), and morphometric scaling improved the fit (Δr = + 0.06). We provide an anatomically grounded and mathematically validated 3D FC model that links microstructures to mesoscale connectivity. Preferential posterior-medial coupling and morphometry-dependent gating support the FC as a modulatory interface in human memory networks and yield testable markers for individualized mapping and clinical translation.

Background: Synthetic neuroimaging data has the potential to augment and improve the generalizability of deep learning models. However, memorization in generative models can lead to unintended leakage of sensitive patient information, limiting model utility and jeopardizing patient privacy.

Methods: We propose RELICT-NI (REpLIca deteCTion-NeuroImaging), a framework for detecting replicas in synthetic neuroimaging datasets. RELICT-NI evaluates image similarity using three complementary approaches: (1) image-level analysis, (2) feature-level analysis via a pretrained medical foundation model, and (3) segmentation-level analysis. RELICT-NI was validated on two clinically relevant neuroimaging use cases: non-contrast head CT with intracerebral hemorrhage (N = 774) and time-of-flight MR angiography of the Circle of Willis (N = 1,782). Expert visual scoring was used as the reference for identifying replicas. Balanced accuracy at the optimal threshold was reported to assess replica classification performance of each method.

Results: The reference visual rating identified 45 of 50 and 5 of 50 generated images as replicas for the NCCT and TOF-MRA use cases, respectively. For the NCCT use case, both image-level and feature-level analyses achieved perfect replica detection (balanced accuracy = 1) at optimal thresholds. A perfect classification of replicas for the TOF-MRA case was not possible at any threshold, with the segmentation-level analysis achieving the highest balanced accuracy (0.79).

Conclusions: Replica detection is a crucial but often neglected validation step in developing deep generative models in neuroimaging. The proposed RELICT-NI framework provides a standardized, easy-to-use tool for replica detection and aims to facilitate responsible and ethical synthesis of neuroimaging data.

Relevance statement: Our developed replica detection framework provides an important step towards standardized and rigorous validation practices of generative models in neuroimaging. Our method promotes the secure sharing of neuroimaging data and facilitates the development of robust deep learning models.

The recent advance in technology allows for the photorealistic digitalization of anatomical specimens that can now be presented through various dynamic visualization modalities, enabling a more interactive learning experience. This study explores a comprehensive workflow for reproducibly integrating photorealistic three-dimensional (3D) anatomical scans of the supra- and infratentorial venous system with stereoscopic visualization and virtual reality (VR) for anatomical learning. A formaldehyde-fixed head and neck specimen was injected with radiopaque dye into its vessels, and a post-mortem computed tomography (CT) venography was performed. A layered anatomical dissection of the intracranial venous system was performed. Photogrammetry surface scanning was employed to create 3D anatomical models, which were then post-processed to produce stereoscopic 3D images and videos using open-source software. In addition, the 3D models were formatted for immersive VR environment integration. Six photorealistic surface models and one CT venography-based reconstruction were generated. These were incorporated into several platforms: multiplayer VR environment using stand-alone headsets, and stereoscopic materials suitable for phone-based VR viewers, 3D multimedia projectors, or 3D monitors with passive or active glasses. These formats supported multiple learning scenarios (VR in single or multiplayer sessions), 3D stereoscopic lectures using 3D multimedia, real-time 3D stereoscopic visualization, or prerecorded videos for phone-based VR visualization. Building on these formats, the proposed workflow enables a realistic and spatially accurate representation of the anatomical data with photorealistic 3D models and facilitates the creation of accessible educational content for 3D stereoscopic presentations and immersive dedicated VR sessions, all through a user-friendly technical approach.

Global interpretability in machine learning holds great potential for extracting meaningful insights from neuroimaging data to improve our understanding of brain function. Although various approaches exist to identify key contributing features at both local and global levels, the high dimensionality and correlations in neuroimaging data require careful selection of interpretability methods to achieve reliable global insights into brain function using machine learning. In this study, we evaluate multiple interpretability techniques such as SHAP, which relies on feature independence, as well as recent advances that account for feature dependence in the context of global interpretability, and inherently global methods such as SAGE. To demonstrate the practical application, we trained XGBoost models to predict age and fluid intelligence using neuroimaging measures from the UK Biobank dataset. By applying these interpretability methods, we found that mean intensities in subcortical regions are consistently and significantly associated with brain aging, while the prediction of fluid intelligence is driven by contributions of the hippocampus and the cerebellum, alongside established regions such as the frontal and temporal lobes. These results underscore the value of interpretable machine learning methods in understanding brain function through a data-driven approach.

Accurate detection of brain midline shift is critical for the diagnosis and monitoring of neurological conditions such as traumatic brain injuries, strokes, and tumors. This study aims to address the lack of dedicated datasets and tools for this task by introducing a novel dataset and a 3D Slicer extension, evaluating the effectiveness of multiple deep learning models for automatic detection of brain midline shift. We introduce the brain-midline-detection dataset, specifically designed for identifying three brain landmarks-Anterior Falx (AF), Posterior Falx (PF), and Septum Pellucidum (SP)-in MRI scans. A comprehensive performance evaluation was conducted using deep learning models including YOLOv5 (n, s, m, l), YOLOv8, and YOLOv9 (GELAN-C model). The best-performing model was integrated into the 3D Slicer platform as a custom extension, incorporating steps such as MRI preprocessing, filtering, skull stripping, registration, and midline shift computation. Among the evaluated models, YOLOv5l achieved the highest precision (0.9601) and recall (0.9489), while YOLOv5m delivered the best mAP@0.5:0.95 score (0.6087). YOLOv5n and YOLOv5s exhibited the lowest loss values, indicating high efficiency. Although YOLOv8s achieved a higher mAP@0.5:0.95 score (0.6382), its high loss values reduced its practical effectiveness. YOLOv9-GELAN-C performed the worst, with the highest losses and lowest overall accuracy. YOLOv5m was selected as the optimal model due to its balanced performance and was successfully integrated into 3D Slicer as an extension for automated midline shift detection. By offering a new annotated dataset, a validated detection pipeline, and open-source tools, this study contributes to more accurate, efficient, and accessible AI-assisted medical imaging for brain midline assessment.

Neuroimaging studies have and continue to advance our understanding of the neurobiology of dyslexia. Integration of data from these studies has the potential to replicate findings, deepen understanding through theoretically focused research, and provide for unexpected discovery. This data integration can be important for questions where a sufficiently large and well-defined group of participants is necessary for sufficient experimental power, particularly for a complex disorder where age, language background, and cognitive profiles can impact imaging results. We have developed a data-sharing platform to provide a data repository, image processing resources, and data analysis tools, with an emphasis on data harmonization across retrospective datasets ( https://dyslexiadata.org ). Here, we summarize data sharing, download, imaging metrics, and quality and privacy considerations in the design of and resources available through this repository. By providing access to a relatively large multisite dataset, researchers can test hypotheses about reading development and disability, test novel data analysis methods, even within the platform, and advance understanding of dyslexia.