Various cognitive disorders have been reported for mild traumatic brain injury (mTBI) patients during the acute stage. This acute stage provides an opportunity for clinicians to optimize treatment protocols, which are based on the evaluation of brain structural connectivity. So far, most brain functional magnetic resonance imaging studies are focused on moderate to severe traumatic brain injuries (TBIs). In this study, we prospectively collected resting state data on 50 mTBI within 3 days of injury and 50 healthy volunteers and analyzed them using Amplitude of low-frequency fluctuation (ALFF), Regional Homogeneity (ReHo), graph theory methods and behavior measure, to explore the dysfunctional brain regions in acute mTBI. In our study, a total of 50 patients suffering <3 days mTBI and 50 healthy subjects were tested in rs-fMRI, as well as under neuropsychological examinations including the Wechsler Intelligence Scale and Stroop Color and Word Test. The correlation analysis was conducted between graph theoretic parameters and neuropsychological results. For the mTBI group, the ReHo of the inferior temporal gyrus and the cerebellum superior are significantly lower than in the control group, and the ALFF of the left insula, the cerebellum inferior, and the middle occipital gyrus were significantly higher than in the control group, which implies the dysfunctionality usually observed in Parkinson's disease. Executive function disorder was significantly correlated with the global efficiencies of the dorsolateral superior frontal gyrus and the anterior cingulate cortex, which is consistent with the literature: the acute mTBI patients demonstrate abnormality in terms of motor speed, association, information processing speed, attention, and short-term memory function. Correlation analysis between the neuropsychological outcomes and the network efficiency for the mTBI group indicates that executive dysfunction might be caused by local brain changes. Our data support the idea that the cerebral internal network has compensatory reactions in response to sudden pathological and neurophysiological changes. In the future, multimode rs-fMRI analysis could be a valuable tool for evaluating dysfunctional brain regions after mTBI.

Brain diseases, including neurodegenerative diseases and neuropsychiatric diseases, have long plagued the lives of the affected populations and caused a huge burden on public health. Functional magnetic resonance imaging (fMRI) is an excellent neuroimaging technology for measuring brain activity, which provides new insight for clinicians to help diagnose brain diseases. In recent years, machine learning methods have displayed superior performance in diagnosing brain diseases compared to conventional methods, attracting great attention from researchers. This paper reviews the representative research of machine learning methods in brain disease diagnosis based on fMRI data in the recent three years, focusing on the most frequent four active brain disease studies, including Alzheimer's disease/mild cognitive impairment, autism spectrum disorders, schizophrenia, and Parkinson's disease. We summarize these 55 articles from multiple perspectives, including the effect of the size of subjects, extracted features, feature selection methods, classification models, validation methods, and corresponding accuracies. Finally, we analyze these articles and introduce future research directions to provide neuroimaging scientists and researchers in the interdisciplinary fields of computing and medicine with new ideas for AI-aided brain disease diagnosis.

Background: For measuring their impact in scientific research, the citation count of the publications is used in the bibliometric analysis, though still in the bibliometric analysis, there is no comprehensive summary of mild traumatic brain injury (mTBI) research. This article intends to provide the physicians and the neuroscientists with a reference guide to assess the most influential publications written on this subject through a macroscopic view of the research activities on mTBI.Methods: The database of the Web of Science was used to compile the 100 top-ranking publications on mTBI. The selected publications were evaluated on the basis of the several categorizations including the type of the publications, number of citations, country of origin, and year of publication.Results: Between 1946 and 2020, the 13,040 publications that were published were included in the database. The least cited publications received 274 citations, while the most cited received 1,748. Altogether, 71 publications were from the USA while 29 were from other countries. Among all the institutions, the University of Pittsburgh Medical Center led the list with six publications. Around 100 papers, mostly on the clinical studies in the categories of neurology and neurosciences, were published in 54 different journals.Conclusions: This study provides both a transverse section summary and historical retrospect for the clinical advances of mTBI, and the publications of important observations that contributed a significant impact on the treatment and prevention of mTBI had been identified.

The convergence of information, communication and data technologies (ICDT) is one of the driving forces of the sixth generation (6G) network. As a typical application of ICDT convergence in the fifth generation (5G) network, the service-based architecture (SBA) has been effectively verified by its large-scale commercial deployment, and the problems are also being gradually solved. At present, the service-based concept has only been applied to the control plane (CP) of core network (CN), and its application on the user plane (UP) of CN is being studied in the industry. For 6G, the service-based radio access network (RAN) should also be put on the agenda to solve problems such as the complexity of interface design, and the long cycle of version upgrades. This paper firstly introduces the driving forces and design principles of service-based RAN. Then it proposes a possible service-based design together with technical advantages. The simulation result finally demonstrates the performance in terms of parallel processing.

In this study, we investigated the feasibility of using electroencephalogram (EEG) signals to differentiate between four distinct subject-driven cognitive states: resting state, narrative memory, music, and subtraction tasks. EEG data were collected from seven healthy male participants while performing these cognitive tasks, and the raw EEG signals were transformed into time–frequency maps using continuous wavelet transform. Based on these time–frequency maps, we developed a convolutional neural network model (TF-CNN-CFA) with a channel and frequency attention mechanism to automatically distinguish between these cognitive states. The experimental results demonstrated that the model achieved an average classification accuracy of 76.14% in identifying these four cognitive states, significantly outperforming traditional EEG signal processing methods and other classical image classification algorithms. Furthermore, we investigated the impact of varying lengths of EEG signals on classification performance and found that TF-CNN-CFA demonstrates consistent performance across different window lengths, indicating its strong generalization capability. This study validates the ability of EEG to differentiate higher cognitive states, which could potentially offer a novel BCI paradigm.

A Network Architecture for Load Balancing of Heterogeneous Wireless Networks