Deep Natural Image Reconstruction from Human Brain Activity Based on Conditional Progressively Growing Generative Adversarial Networks / 神经科学通报·英文版

Brain decoding based on functional magnetic resonance imaging has recently enabled the identification of visual perception and mental states. However, due to the limitations of sample size and the lack of an effective reconstruction model, accurate reconstruction of natural images is still a major challenge. The current, rapid development of deep learning models provides the possibility of overcoming these obstacles. Here, we propose a deep learning-based framework that includes a latent feature extractor, a latent feature decoder, and a natural image generator, to achieve the accurate reconstruction of natural images from brain activity. The latent feature extractor is used to extract the latent features of natural images. The latent feature decoder predicts the latent features of natural images based on the response signals from the higher visual cortex. The natural image generator is applied to generate reconstructed images from the predicted latent features of natural images and the response signals from the visual cortex. Quantitative and qualitative evaluations were conducted with test images. The results showed that the reconstructed image achieved comparable, accurate reproduction of the presented image in both high-level semantic category information and low-level pixel information. The framework we propose shows promise for decoding the brain activity.

A Longitudinal Functional Magnetic Resonance Imaging Study of Working Memory in Patients Following a Transient Ischemic Attack: A Preliminary Study / 神经科学通报·英文版

In this study, we used functional magnetic resonance imaging (fMRI) to investigate longitudinal changes in brain activation during a verbal working memory (VWM) task performed by patients who had experienced a transient ischemic attack (TIA). Twenty-five first-ever TIA patients without visible lesions in conventional MRI and 25 healthy volunteers were enrolled. VWM task-related fMRI was conducted 1 week and 3 months post-TIA. The brain activity evoked by the task and changes over time were assessed. We found that, compared with controls, patients exhibited an increased activation in the bilateral inferior frontal gyrus (IFG), right dorsolateral prefrontal cortex (DLPFC), insula, inferior parietal lobe (IPL), and cerebellum during the task performed 1 week post-TIA. But only the right IFG still exhibited an increased activation at 3 months post-TIA. A direct comparison of fMRI data between 1 week and 3 months post-TIA showed greater activation in the bilateral middle temporal gyrus, right DLPFC, IPL, cerebellum, and left IFG in patients at 1 week post-TIA. We conclude that brain activity patterns induced by a VWM task remain dynamic for a period of time after a TIA, despite the cessation of clinical symptoms. Normalization of the VWM activation pattern may be progressively achieved after transient episodes of ischemia in TIA patients.

Analysis of Altered Baseline Brain Activity in Drug-Naive Adult Patients with Social Anxiety Disorder Using Resting-State Functional MRI

OBJECTIVE: We hypothesize that the amplitude of low-frequency fluctuations (ALFF) is involved in the altered regional baseline brain function in social anxiety disorder (SAD). The aim of the study was to analyze the altered baseline brain activity in drug-naive adult patients with SAD. METHODS: We investigated spontaneous and baseline brain activities by obtaining the resting-state functional magnetic resonance imaging data of 20 drug-naive adult SAD patients and 19 healthy controls. Voxels were used to analyze the ALFF values using one- and two-sample t-tests. A post-hoc correlation of clinical symptoms was also performed. RESULTS: Our findings show decreased ALFF in the bilateral insula, left medial superior frontal gyrus, left precuneus, left middle temporal gyrus, right middle temporal pole, and left fusiform gyrus of the SAD group. The SAD patients exhibited significantly increased ALFF in the right inferior temporal gyrus, right middle temporal gyrus, bilateral middle occipital gyrus, orbital superior frontal gyrus, right fusiform gyrus, right medial superior frontal gyrus, and left parahippocampal gyrus. Moreover, the Liebowitz Social Anxiety Scale results for the SAD patients were positively correlated with the mean Z values of the right middle occipital and right inferior occipital but showed a negative correlation with the mean Z values of the right superior temporal gyrus and right medial superior frontal gyrus. CONCLUSION: These results of the altered regional baseline brain function in SAD suggest that the regions with abnormal spontaneous activities are involved in the underlying pathophysiology of SAD patients.

A brain functional connectivity analysis based on independent component analysis / 生物医学工程学杂志

The resting state cortical functional connectivity is an important method in current brain researches. In this paper, we propose an approach for analyzing and manipulating the resting state functional magnetic resonance imaging (fMRI) data using spatial independent component analysis (sICA) method, and applying the low-frequency oscillations theory to the choice of component of interest (COI) from the component obtained by sICA method. Firstly, we remove all the inactive voxels and independent voxels via Z value. Then, by making a spectrum analysis, we choose the COI with concentrations of energy between 0.01 and 0.1 Hz. And after that, we obtain the functional connectivity networks using hierarchical clustering.

Electroencephalogram-functional magnetic resonance imaging study of the interictal epileptic activity in two patients with partial epilepsy / 中华神经科杂志

Objective To investigate the map and pattern of blood oxygen level dependent(BOLD)signal changes correlated to interictal epileptiform discharges(IEDs)with EEG-fMRI in patients with partial epilepsy and then to explore the pathophysiological mechanisms of epileptic discharges and their effect on brain function in partial epilepsy.Methods Through the method of EEG-fMRI,2 patients with parial epilepsy were studied.The relationship between the regions of BOLD signal changes linked to IEDs and the electroelinical localization of epileptogenic zone in patients with partial epilepsy were investigated.Results The epileptogenic areas localized by electroclinical findings in the 2 patients all showed maximal activation and 2 sites of significant activation were found in 1 of the 2 patients;Weak activation were also manifested in the opposite side corresponding to lesions.Conclusions IED-linked BOLD response in patients with partial epilepsy is mainly in epileptogenic zones and weak activation can also be seen in the corresponding contralateral areas of epileptogenic zoiles.Activation areas ale well concordant with epileptogenie areas localized by electroclinical findings.

A new method based on sparse component decomposition to remove MRI artifacts in the continuous EEG recordings / 生物医学工程学杂志

How to effectively remove the magnetic resonance imaging (MRI) artifacts in the electroencephalography (EEG) recordings, when EEG and functional magnetic resonance imaging (FMRI) are simultaneous recorded, is a challenge for integration of EEG and FMRI. According to the temporal-spatial difference between MRI artifacts and EEG, a new method based on sparse component decomposition in the mixed over-complete dictionary is proposed in this paper to remove MR artifacts. A mixed over-complete dictionary (MOD) of waveletes and discrete cosine which can exhibit the temporal-spatial discrepancy between MRI artificats and EEG is constructed first, and then the signals are separated by learning in this MOD with matching pursuit (MP) algorithm. The method is applied to the MRI artifacts corrupted EEG recordings and the decomposition result shows its validation.

Advances in independent component analysis and its application / 生物医学工程学杂志

The independent component analysis (ICA) is a new technique in statistical signal processing, which decomposes mixed signals into statistical independent components. The reported applications in biomedical and radar signal have demonstrated its good prospect in various blind signal separation. In this paper, the progress of ICA in such as its principle, algorithm and application and advance direction of ICA in future is reviewed. The aim is to promote the research in theory and application in the future.

A method based on independent component analysis for processing fMRI data / 生物医学工程学杂志

Independent component analysis (ICA) is a new technique in statistical signal processing to extract independent components from multidimensional measurements of mixed signals. In this paper, for the processing of functional magnetic resonance imaging(fMRI) data, two signals of near voxels are used as the mixed signals and are separated by ICA. The correlation coefficients between the reference signal and the separated signals are calculated and those voxels whose correlation coefficients are greater than a threshold are considered to be the activated voxels by the stimulation, and so the functional localization of the stimulation is completed. The validity of the method was primarily proved by trial of real brain functional magnetic resonance imaging data.