Neuroscience informatics最新文献

Objective

Endovascular embolization has an important role in the management of brain arteriovenous malformations (AVMs). A Tsinghua AVM grading system has been proposed for patient selection and complete obliteration. The authors sought to validate this system in an independent patient cohort and compare it to the Buffalo grading system.

Methods

Consecutive 52 patients underwent endovascular AVM embolization between January 2019 and December 2021 according to Tsinghua AVM grading system. Each AVM was also graded using Buffalo grading system. Baseline clinical characteristics, complications, and AVM obliteration were compared between Tsinghua and Buffalo scales.

Results

Complete obliteration of AVM was obtained in 29 patients (55.8%). Three complications were encountered, one bleeding (1.9%) and 2 ischemic (3.8%), in 3(5.7%) patients who recovered completely at follow-up. The Tsinghua scale (p=0.017) was predictor of complete obliteration as well as Buffalo scale (p=0.002) on ROC curve analysis and their AUCs were not significantly different (p=0.672). The Tsinghua scale was also associated with the initial patient status (p=0.003) and injected Onyx volume (p=0.003) on linear regression test. Because of the low complication rate, neither the Tsinghua scale nor the Buffalo scale predicted complication risk related to AVM embolization.

Conclusions

The bleeding complication rate of 1.9% is within the range of rupture risk reported in the natural history of AVMs. In addition to predicting complete AVM obliteration as well as Buffalo scale, the Tsinghua scale can also predict the patients' status and the volume of Onyx avoid over injection.

Key messages

The Tsinghua grading system for endovascular AVM embolization will guide patient selection of AVM embolization.

Cognitive neuroscience investigates the intricate connections between brain function and mental processing to understand the cognitive architecture. Exploring the human brain, the epicenter of cognitive activity, offers valuable insights into underlying cognitive processes. To monitor brain states corresponding to various mental activities, appropriate measurement tools are essential. Electroencephalogram (EEG) signals serve as a valuable tool for recording patterns and changes in electrical brain activities. Leveraging non-linear signal processing techniques holds promise for advancing our understanding of brain activities during cognitive tasks. In this study, we analyze the electrical activity of the brain using EEG data collected from subjects engaged in a cognitive workload task. Employing wavelet-based analysis, we capture changes in the structure of EEG signals before and during a mental arithmetic task. Additionally, spectral analysis is conducted to discern alterations in the distribution of spectral contents of EEG signals. Our findings underscore the efficacy of wavelet-based analysis and spectral entropy in quantifying the time-varying and non-stationary nature of EEG recordings, offering effective frameworks for distinguishing between different cognitive activities. Consequently, these methods afford deeper insights into the cognitive architecture by tracking changes in the distribution of the time-varying spectrum.

Background and Purpose

With an increased need for standardized methodology in accurate diagnosis of Traumatic Brain Injury (TBI), Diffusion Tensor Imaging (DTI) has provided promising diagnostic results as an adjunct modality yet remains underutilized. The purpose of this study was to validate the use of DTI with Statistical Parametric Mapping (SPM) for Traumatic Brain Injury (TBI) supporting its use as a diagnostic tool.

Materials and Methods

This study was retrospective and compared controls to patients clinically diagnosed with TBI. Forty-two controls (mean age = 34.1; range, 19 - 58; 28 Males and 13 Females) were screened (n = 41) for cognitive impairment and neurological abnormality. Two cohorts, each of eighteen patients (first cohort: mean age, 41.8; range, 23 - 70; 9 Males and 9 Females; second cohort: mean age, 45.7; range, 23 - 68; 9 Males and 9 Females) clinically diagnosed with TBI (n = 36) were pooled. DTI image acquisition was obtained using a 3 Tesla MRI scanner. DTI images were analyzed through voxel-based t-tests using SPM comparing each individual to the normative control group to generate z-maps for each individual control and each individual patient with a TBI. Test statistics were ranged for p-values (0.001-0.05) and cluster extent values (0, 30, 60, 65, 70, 75). Area Underneath A Receiver Operating Characteristic Curve (AUCROC) was used to validate diagnostic capability. AUCROC analysis was conducted on all sets of p-value and extent threshold values. Significance of results was determined by examining the AUCROC values.

Results and Conclusions

A maximal AUCROC of 1.000 was obtained across the p-value range and cluster extent thresholding values specified across the two cohorts. The high AUCROC supports validation of the methodology presented and the use of diffusion tensor imaging and SPM as an adjunct diagnostic tool for TBI.

Automated brain tumor segmentation from magnetic resonance (MR) images plays a crucial role in precise diagnosis and treatment monitoring in brain tumor care. Leveraging the Brain Tumor Segmentation Challenge (BraTS) dataset, this paper introduces an extended version of the nnU-Net architecture for brain tumor segmentation, addressing both adult (Glioma) and pediatric tumors. Our methodology integrates innovative approaches to enhance segmentation accuracy. We incorporate residual blocks to capture complex spatial features, attention gates to emphasize informative regions and implement the Hausdorff distance (HD) loss for boundary-based segmentation refinement. The effectiveness of each enhancement is systematically evaluated through an ablation study using different configurations on the BraTS dataset. Results from our study highlight the significance of combining residual blocks, attention gates, and HD loss, achieving the best performance with a mean Dice and HD score of 83%, 3.8 and 71%, and 8.7 for Glioma and Pediatrics datasets, respectively. This advanced nnU-Net showcases the promising potential for accurate and robust brain tumor segmentation, offering valuable insights for enhanced clinical decision-making in pediatric brain tumor care.

Introduction

While linguistic retrogenesis has been extensively investigated in the neuroscientific and behavioral literature, there has been little work on retrogenesis using computerized approaches to language analysis.

Methods

We bridge this gap by introducing a method based on comparing output of a pre-trained neural language model (NLM) with an artificially degraded version of itself to examine the transcripts of speech produced by seniors with and without dementia and healthy children during spontaneous language tasks. We compare a range of linguistic characteristics including language model perplexity, syntactic complexity, lexical frequency and part-of-speech use across these groups.

Results

Our results indicate that healthy seniors and children older than 8 years share similar linguistic characteristics, as do dementia patients and children who are younger than 8 years.

Discussion

Our study aligns with the growing evidence that language deterioration in dementia mirrors language acquisition in development using computational linguistic methods based on NLMs. This insight underscores the importance of further research to refine its application in guiding developmentally appropriate patient care, particularly in early stages.

EEGLAB is one of the most famous software for processing, analyzing, and researching experiments that have Electroencephalography (EEG) datasets. Due to the numerous add-ins along with global, widespread communications and online free YouTube channel, its popularity increased every year. To address this phenomenon from a bibliographic perspective, we found 20,464 citations in Google Scholar since 8/27/2023. Then, only the Web of Science (WOS) articles were 12,700 that they were extracted. The results were analyzed with Bibliometrix package from CRAN R software. The time span of these articles is from 2004 to 2023 with 12,700 documents in 1,125 sources (journals, books, etc.), 29,125 authors, 19,062 author's keywords, 13,707 keywords PLUS, 279,617 references. The annual growth rate is 28.12%, international Co-authorship is 37.27%, Co-authors per document is 4.89 and the average citations per document is 22.51. The most relevant sources are Neuroimage, Frontiers in Human Neurosciences, Scientific Reports, Psychophysiology, and PLOS One with 780, 526, 446,425, and 371 articles, respectively. The most cited countries are the USA, Germany, and the United Kingdom with 93,093, 32,621, and 20,748 total citations, respectively. The ERPLAB, ADJUST, and ICLabel add-ins have the local to global citation ratios equal to 85.4%, 65.1%, and 78.2% respectively. The collaboration network university, trend topic plot of keyword plus, thematic map trigram word in abstract and co-citation network of published papers after 2018 are presented. EEGLAB is among the most cited MATLAB toolboxes in computational neuroscience. Many developed and developing countries use it in their research publications.

Recently, characterizing the dynamics of brain functional networks at task free or cognitive tasks has developed different research efforts in the field of neuroscience. Epilepsy is an electrophysiological brain disease which is accompanied by recurrent seizures. Seizure and epilepsy detection is a main challenge in the field of neuroscience. Understanding the underlying mechanism of epilepsy and transition from a normal brain to epileptic brain crucial for the diagnosis and treatment purposes. To understand the organization of epileptic brain network functions at large scales, electroencephalogram (EEG) signals measure and record the changes in electrical activity and functional connectivity. Time frequency analysis and continuous spectral entropy are well developed methods which reveal dynamical aspects of brain activity and can analyze the transitions in intrinsic brain activity. In this work, we aim to model the dynamics of EEG signals of epileptic brain and characterize their dynamical patterns. We use Time frequency analysis to capture the alterations in the structure of EEG signals from patients with seizure. Continuous spectral entropy is used to detect the start of seizures. The main purpose of the current is to explore the changes in the organization of epileptic brain networks. Using time frequency techniques, we are able to draw a big picture of how the brain functions before and during seizure and step forward to classify seizure and corresponding brain activity during different stages of epilepsy. The present study may contribute to characterizing the complex non-linear dynamics of EEG signals of epileptic brain and further assists with biomarker detection for different clinical applications. This finding helps towards effective diagnosis and better treatment of epilepsy.

Epilepsy is the neurological disorder affecting around 50 million people worldwide. It is characterized by recurrent and unpredictable seizures. Correctly counting seizure occurrences is crucial for diagnosis and treatment of epilepsy, which will lower the risk of SUDEP (sudden unexpected deaths in epilepsy). Many previous researches on patient-specific seizure detection have obtained a good performance but with limited practicability in clinical setting. On the other hand, patient non-specific detection is clinically practicable but with limited performance. This study aims to improve the performance of patient non-specific seizure detection by comparing performances among one modality based models and multi-modal based model. The study was based on clinical data from the open source Siena Scalp EEG Database, which consist of simultaneous EEG (Electroenchephalography) and ECG (electrocardiography) recording from 14 patients with focal epilepsy. The seizures were annotated by an epilepsy expert after a careful review of the clinical and EEG data of each patient. First, relevant signal pre-processing were performed, followed by features extraction. Then, machine learning approach based on random forest was employed for seizure detection with leave-one-patient-out cross validation scheme. EEG detector and ECG detector were separately trained with each signal. Multi-modal detector was based on combining EEG detector and ECG detector by the late integration approach with the Boolean operation “OR” strategy. The performances were compared among those three detectors and with the state of the art. The result has shown that the multi-modal detector achieved a sensitivity of 87.62% and outperformed the ECG detector (41.55%), the EEG detector (81.43%), and the state-of-the-art non-specific detectors. Notably, the ECG detector detected some seizures which EEG detector failed to detect. This indicated that the ECG signal was beneficial for increasing sensitivity. However, due to the “OR” fusion strategy, the multi-modal detector also inherited the false detections resulted from either EEG detector or ECG detector. The findings of the study demonstrate the potential of improving performance of patient non-specific seizure detection by multimodal data. It shows that the proposed method should be further validated on large database and further development should focus on lowering false detections before clinical application.

Recent advances in brain network analysis are largely based on graph theory methods to assess brain network organization, function, and malfunction. Although, functional magnetic resonance imaging (fMRI) has been frequently used to study brain activity, however, the nonlinear dynamics in resting-state (fMRI) data have not been extensively characterized. In this work, we aim to model the dynamics of resting-state (fMRI) and characterize the dynamical patterns in resting-state (fMRI) time series data in left and right hippocampus and inferior frontal gyrus. We use Sliding Window Embedding (SWE) method to reconstruct the phase space of resting-state (fMRI) data from left and right hippocampus and orbital part of inferior frontal gyrus. The complexity of resting-state MRI data is examined using fractal analysis. The main purpose of the current study is to explore the topological organization of hippocampus and frontal gyrus and consequently, memory. By constructing resting-state functional network from resting-state (fMRI) time series data, we are able to draw a big picture of how brain functions and step forward to classify brain activity between normal control people and patients with different brain disorders.