Toward Understanding and Boosting Adversarial Transferability From a Distribution Perspective.

Transferable adversarial attacks against Deep neural networks (DNNs) have received broad attention in recent years. An adversarial example can be crafted by a surrogate model and then attack the unknown target model successfully, which brings a severe threat to DNNs. The exact underlying reasons for the transferability are still not completely understood. Previous work mostly explores the causes from the model perspective, e.g., decision boundary, model architecture, and model capacity. Here, we investigate the transferability from the data distribution perspective and hypothesize that pushing the image away from its original distribution can enhance the adversarial transferability. To be specific, moving the image out of its original distribution makes different models hardly classify the image correctly, which benefits the untargeted attack, and dragging the image into the target distribution misleads the models to classify the image as the target class, which benefits the targeted attack. Towards this end, we propose a novel method that crafts adversarial examples by manipulating the distribution of the image. We conduct comprehensive transferable attacks against multiple DNNs to demonstrate the effectiveness of the proposed method. Our method can significantly improve the transferability of the crafted attacks and achieves state-of-the-art performance in both untargeted and targeted scenarios, surpassing the previous best method by up to 40% in some cases. In summary, our work provides new insight into studying adversarial transferability and provides a strong counterpart for future research on adversarial defense.

Acellular dermal matrix decorated with collagen-affinity peptide accelerate diabetic wound healing through sustained releasing Histatin-1 mediated promotion of angiogenesis.

Treating diabetic ulcers is a major challenge in clinical practice, persecuting millions of patients with diabetes and increasing the medical burden. Recombinant growth factor application can accelerate diabetic wound healing via angiogenesis. The local administration of recombinant growth factors has no robust clinical efficiency because of the degradation of append short duration of the molecules in the hostile inflammatoryenvironment.The present study focused on the pathophysiology of impaired neovascularization and growth factor short duration in the diabetic wound. We prepared a collagen-binding domain (CBD)-fused recombinant peptide (C-Histatin-1) that had both pro-angiogenesis capacity and collagen-affinity properties. Next, we created a biocompatible acellular dermal matrix (ADM) as a drug delivery carrier that featured collagen-richness, high porosity, and non-cytotoxicity. C-Histatin-1 was then tethered on ADM to obtain a sustained-release effect. Finally, a functional scaffold (C-Hst1/ADM) was developed. C-Hst1/ADM can sustain-release Histatin-1 to promote the adhesion, migration, and angiogenesisof vascular endothelial cells in vitro. Using a diabetic wound model, we showed that C-Hst1/ADM could significantly promote angiogenesis, reduce scar widths, and improve extracellular collagen accumulation. Therefore, the results of this study provide a foundation for the clinical application of C-Hst1/ADM covering scaffold in the treatment of diabetic wounds.

TRIM21 inhibits the osteogenic differentiation of mesenchymal stem cells by facilitating K48 ubiquitination-mediated degradation of Akt.

Tripartite motif containing 21 (TRIM21) is a member of the TRIM protein family with E3 ubiquitin ligase activity. Recent studies have demonstrated that TRIM21 widely contributes to physiological and pathological processes by ubiquitylating critical proteins in many kinds of cells. Additionally, multiple studies have shown that TRIM21 plays an important role in multiple cell differentiation processes. However, whether TRIM21 modulates the osteogenic differentiation process of mesenchymal stem cells (MSCs) remains unclear. In this study, we demonstrated that the expression of TRIM21 was decreased during the osteogenic process of MSCs and that TRIM21 negatively regulated the osteogenic capacity of MSCs both in vitro and in vivo. Moreover, we further demonstrated that TRIM21 modulated the osteogenic process of MSCs by acting as an E3 ubiquitin ligase to mediate the K48-linked ubiquitination of Akt and cause degradation. In summary, these results emphasize the critical role of TRIM21 in bone formation and TRIM21 may be a promising target to improve the clinical use of MSCs in tissue engineering.

A Feature Extraction Method for Seizure Detection Based on Multi-Site Synchronous Changes and Edge Detection Algorithm.

Automatic detection of epileptic seizures is important in epilepsy control and treatment, and specific feature extraction assists in accurate detection. We developed a feature extraction method for seizure detection based on multi-site synchronous changes and an edge detection algorithm. We investigated five chronic temporal lobe epilepsy rats with 8- and 12-channel detection sites in the hippocampus and limbic system. Multi-site synchronous changes were selected as a specific feature and implemented as a seizure detection method. For preprocessing, we used magnitude-squared coherence maps and Canny edge detection algorithm to find the frequency band with the most significant change in synchronization and the important channel pairs. In detection, we used the maximal cross-correlation coefficient as an indicator of synchronization and the correlation coefficient curves' average value and standard deviation as two detection features. The method achieved high performance, with an average 96.60% detection rate, 2.63/h false alarm rate, and 1.25 s detection delay. The experimental results show that synchronization is an appropriate feature for seizure detection. The magnitude-squared coherence map can assist in selecting a specific frequency band and channel pairs to enhance the detection result. We found that individuals have a specific frequency band that reflects the most significant synchronization changes, and our method can individually adjust parameters and has good detection performance.

[Comparison of pulse oxygen saturation/fraction of inhaled oxygen and arterial partial pressure of oxygen/fraction of inhaled oxygen in the assessment of oxygenation in acute respiratory distress syndrome patients at different high altitudes in Yunnan Province].

OBJECTIVE: To investigate and evaluate if pulse oxygen saturation/fraction of inhaled oxygen (SpO2/FiO2) can be used, as replacement of arterial partial pressure of oxygen/fraction of inhaled oxygen (PaO2/FiO2), to assess oxygenation in acute respiratory distress syndrome (ARDS) patients at different high altitudes in Yunnan Province, and to find a rapid and non-invasive method for the diagnosis of ARDS at different altitudes. METHODS: Patients with ARDS at different high altitudes in Yunnan Province from January 2019 to December 2020 were enrolled. The patients were divided into three groups according to different altitudes, and received different oxygen therapies according to their respective medical conditions. Group 1 consisted of patients with moderate to severe ARDS from the department of critical care medicine of the First Affiliated Hospital of Kunming Medical University (average altitude approximately 1 800 m), and received mechanical ventilation to maintain SpO2 of 0.90-0.96 with a low FiO2 for more than 30 minutes, and SpO2, FiO2, PaO2 were recorded. Group 2 consisted of patients with moderate to severe ARDS at the department of critical care medicine of People's Hospital of Diqing Tibetan Autonomous Prefecture (mean altitude about 3 200 m), and received oxygen with an attached reservoir mask to maintain SpO2 of 0.90-0.96 for 10 minutes, and then SpO2, FiO2, and PaO2 were recorded. Group 3 consisted of patients with mild to moderate-severe ARDS who admitted to the emergency department of the People's Hospital of Lijiang (average altitude approximately 2 200 m); when SpO2 < 0.90, patients received oxygen with the oxygen storage mask, and the FiO2 required to maintain SpO2 ≥ 0.90 was recorded, and SpO2, FiO2, PaO2 were recorded after oxygen inhalation for 10 minutes. Spearman coefficient was used to analyze the correlation between SpO2/FiO2 and PaO2/FiO2 in each group. Linear analysis was used to derive the linear equation between SpO2/FiO2 and PaO2/FiO2, and to evaluate arterial pH, arterial partial pressure of carbon dioxide (PaCO2), FiO2, tidal volume (VT), positive end-expiratory pressure (PEEP) and other related factors which would change the correlation between SpO2/FiO2 and PaO2/FiO2. The receiver operator characteristic curve (ROC curve) was plotted to calculate the sensitivity and specificity of using SpO2/FiO2 instead of PaO2/FiO2 to assess oxygenation of ARDS patients. RESULTS: Group 1 consisted of 24 ARDS patients from whom 271 blood gas analysis results were collected; group 2 consisted of 14 ARDS patients from whom a total of 47 blood gas analysis results were collected; group 3 consisted of 76 ARDS patients, and a total of 76 blood gas analysis results were collected. The PaO2/FiO2 (mmHg, 1 mmHg = 0.133 kPa) in groups 1, 2 and 3 were 103 (79, 130), 168 (98, 195) and 232 (146, 271) respectively, while SpO2/FiO2 were 157 (128, 190), 419 (190, 445) and 319 (228, 446) respectively. Among the three groups, patients in group 1 had the lowest PaO2/FiO2 and SpO2/FiO2, while patients in group 3 had the highest. Spearman correlation analysis showed that PaO2/FiO2 was highly correlated with SpO2/FiO2 in groups 1, 2 and 3 (r values were 0.830, 0.951, 0.828, all P < 0.05). Regression equation was fitted according to linear analysis: in group 1 SpO2/FiO2 = 58+0.97×PaO2/FiO2 (R2 = 0.548, P < 0.001); in group 2 SpO2/FiO2 = 6+2.13×PaO2/FiO2 (R2 = 0.938, P < 0.001); in group 3 SpO2/FiO2 = 53+1.33×PaO2/FiO2 (R2 = 0.828, P < 0.001). Further analysis revealed that PEEP, FiO2, and arterial blood pH could affect the correlation between SpO2/FiO2 and PaO2/FiO2. ROC curve analysis showed that the area under ROC curve (AUC) was 0.848 and 0.916 in group 1 with moderate to severe ARDS; based on the regression equation, the corresponding SpO2/FiO2 cut-off values at a PaO2/FiO2 of 100 mmHg and 200 mmHg were 155, 252 with a sensitivity of 84.9% and 100%, specificity of 87.2% and 70.6%, respectively. Patients with moderate to severe ARDS in group 2 (AUC was 0.945 and 0.977), the corresponding SpO2/FiO2 cut-off values at PaO2/FiO2 of 100 mmHg and 200 mmHg were 219 and 432 with the sensitivity of 100% and 85.2%, specificity of 82.5% and 100%, respectively. Patients with mild to moderate-severe ARDS in group 3 (AUC was 0.903 and 0.936), the corresponding SpO2/FiO2 cut-off values at a PaO2/FiO2 of 200 mmHg and 300 mmHg were 319 and 452 with the sensitivity of 100% and 100%, specificity of 80.9% and 86.2%, respectively. CONCLUSIONS: SpO2/FiO2 and PaO2/FiO2 in ARDS patients at different high altitudes in Yunnan Province have a good correlation, and non-invasive SpO2/FiO2 can be used to replace PaO2/FiO2 to assess the oxygenation in ARDS patients.

Pneumonia Detection Using an Improved Algorithm Based on Faster R-CNN.

Pneumonia remains a threat to human health; the coronavirus disease 2019 (COVID-19) that began at the end of 2019 had a major impact on the world. It is still raging in many countries and has caused great losses to people's lives and property. In this paper, we present a method based on DeepConv-DilatedNet of identifying and localizing pneumonia in chest X-ray (CXR) images. Two-stage detector Faster R-CNN is adopted as the structure of a network. Feature Pyramid Network (FPN) is integrated into the residual neural network of a dilated bottleneck so that the deep features are expanded to preserve the deep feature and position information of the object. In the case of DeepConv-DilatedNet, the deconvolution network is used to restore high-level feature maps into its original size, and the target information is further retained. On the other hand, DeepConv-DilatedNet uses a popular fully convolution architecture with computation shared on the entire image. Then, Soft-NMS is used to screen boxes and ensure sample quality. Also, K-Means++ is used to generate anchor boxes to improve the localization accuracy. The algorithm obtained 39.23% Mean Average Precision (mAP) on the X-ray image dataset from the Radiological Society of North America (RSNA) and got 38.02% Mean Average Precision (mAP) on the ChestX-ray14 dataset, surpassing other detection algorithms. So, in this paper, an improved algorithm that can provide doctors with location information of pneumonia lesions is proposed.

Adipose-Derived Stromal Cell-Sheets Sandwiched, Book-Shaped Acellular Dermal Matrix Capable of Sustained Release of Basic Fibroblast Growth Factor Promote Diabetic Wound Healing.

The management of diabetic wounds is a therapeutic challenge in clinical settings. Current tissue engineering strategies for diabetic wound healing are insufficient, owing to the lack of an appropriate scaffold that can load a large number of stem cells and induce the interaction of stem cells to form granulation tissue. Herein we fabricated a book-shaped decellularized dermal matrix (BDDM), which shows a high resemblance to native dermal tissue in terms of its histology, microstructure, and ingredients, is non-cytotoxic and low-immunogenic, and allows adipose-derived stromal cell (ASC) attachment and proliferation. Then, a collagen-binding domain (CBD) capable of binding collagen was fused into basic fibroblast growth factor (bFGF) to synthetize a recombinant growth factor (termed as CBD-bFGF). After that, CBD-bFGF was tethered onto the collagen fibers of BDDM to improve its endothelial inducibility. Finally, a functional scaffold (CBD-bFGF/BDDM) was fabricated. In vitro and in vivo experiments demonstrated that CBD-bFGF/BDDM can release tethered bFGF with a sustained release profile, steadily inducing the interaction of stem cells down to endothelial differentiation. ASCs were cultured to form a cell sheet and then sandwiched by CBD-bFGF/BDDM, thus enlarging the number of stem cells loaded into the scaffold. Using a rat model, the ASC sheets sandwiched with CBD-bFGF/BDDM (ASCs/CBD-bFGF/BDDM) were capable of enhancing the formation of granulation tissue, promoting angiogenesis, and facilitating collagen deposition and remodeling. Therefore, the findings of this study demonstrate that ASCs/CBD-bFGF/BDDM could be applicable for diabetic wound healing.

[Feasibility of pulse oxygen saturation for the condition evaluation of acute respiratory distress syndrome patients in the Lijiang region].

OBJECTIVE: To explore the feasibility of using pulse oxygen saturation (SpO2) to evaluate the condition of patients with acute respiratory distress syndrome (ARDS) in the Lijiang region. METHODS: Patients with ARDS who visited the department of emergency of People's Hospital of Lijiang from August to December 2020 were selected as study subjects. Patients were divided by severity into mild ARDS group [200 mmHg (1 mmHg = 0.133 kPa) ≤ oxygenation index (PaO2/FiO2, P/F) ≤ 300 mmHg] and moderate to severe ARDS group (P/F ≤ 200 mmHg). The general condition, clinical diagnosis, arterial blood gas analysis results of the patients were recorded, and the differences of the above indexes between the two groups of ARDS were compared. Spearman correlation analysis was used to analyze the correlation between SpO2 and arterial oxygen saturation (SaO2). SpO2 was carried into the Ellis equation and the Rice equation to calculate the derived P/F and analyze the correlation between the derived P/F and the P/F measured in arterial blood gas analysis; receiver operator characteristic curve (ROC curves) were plotted, the sensitivity and specificity of SpO2/fraction of inspiration oxygen (SpO2/FiO2, S/F) instead of P/F to assess oxygenation in patients with ARDS was calculated. To evaluate the feasibility of SpO2 for the condition evaluation of patients with ARDS in the Lijiang region. RESULTS: Compared with the mild ARDS group, the arterial partial pressure of oxygen (PaO2), SaO2 and hemoglobin (Hb) were significantly decreased in the moderate to severe ARDS group [PaO2 (mmHg): 50.5 (39.3, 56.5) vs. 60.0 (55.0, 67.5), SaO2: 0.86 (0.73, 0.91) vs. 0. 93 (0.90, 0.96), Hb (g/L): 142±27 vs. 156±24, respectively, all P < 0.05]. Correlation analysis revealed a significant positive correlation between SpO2 and SaO2 in ARDS patients residing at high altitude (R = 0.650, P = 0.000). The P/F derived by the Rice formula was significantly and positively correlated with the P/F derived from arterial blood gas analysis (R = 0.802, P = 0.000). The deduced P/F in mild and moderate to severe ARDS groups were all significantly correlated with the measured P/F (R values were 0.562, 0.647, both P = 0.000). The P/F derived using the Ellis formula showed a significant positive correlation with the P/F derived from arterial blood gas analysis (R = 0.822, P = 0.000). The deduced P/F of mild ARDS group and moderate to severe ARDS group were all positively correlated with the measured P/F (R values were 0.556, 0.589, P values were 0.000, 0.010). There was a significant positive correlation between S/F and P/F in ARDS patients (R = 0.828, P = 0.000), and the regression equation was S/F = 1.33 P/F+52.41. ROC curve analysis showed that S/F had some predictive value for patients with mild and moderate to severe ARDS, and area under ROC curve (AUC) and 95% confidence interval (95%CI) were 0.903 (0.829-0.977), 0.936 (0.870-1.000), both P = 0.000. When the cut-off value was 452 mmHg, S/F had a sensitivity of 100% and a specificity of 80.9% for predicting mild ARDS. When the cut-off value was 319 mmHg, S/F predicted moderate to severe ARDS with 95.1% sensitivity and 86.2% specificity. CONCLUSIONS: At high altitude, SpO2 and SaO2 have been correlated in patients with ARDS, and P/F derived using SpO2 and measured P/F were significantly correlated in patients with ARDS, especially in those with moderate to severe ARDS. SpO2 may be useful in the assessment of severity of illness in patients with ARDS at high altitude.

Optimization of Sparse Cross Array Synthesis via Perturbed Convex Optimization.

Three-dimensional (3-D) imaging sonar systems require large planar arrays, which incur hardware costs. In contrast, a cross array consisting of two perpendicular linear arrays can also support 3-D imaging while dramatically reducing the number of sensors. Moreover, the use of an aperiodic sparse array can further reduce the number of sensors efficiently. In this paper, an optimized method for sparse cross array synthesis is proposed. First, the beamforming of a cross array based on a multi-frequency algorithm is simplified for both near-field and far-field. Next, a perturbed convex optimization algorithm is proposed for sparse cross array synthesis. The method based on convex optimization utilizes a first-order Taylor expansion to create position perturbations that can optimize the beam pattern and minimize the number of active sensors. Finally, a cross array with 100 + 100 sensors is employed from which a sparse cross array with 45 + 45 sensors is obtained via the proposed method. The experimental results show that the proposed method is more effective than existing methods for obtaining optimum results for sparse cross array synthesis in both the near-field and far-field.

N-Doped Modified Graphene/Fe2O3 Nanocomposites as High-Performance Anode Material for Sodium Ion Storage.

Sodium-ion storage devices have received widespread attention because of their abundant sodium resources, low cost and high energy density, which verges on lithium-ion storage devices. Electrochemical redox reactions of metal oxides offer a new approach to construct high-capacity anodes for sodium-ion storage devices. However, the poor rate performance, low Coulombic efficiency, and undesirable cycle stability of the redox conversion anodes remain a huge challenge for the practical application of sodium ion energy storage devices due to sluggish kinetics and irreversible structural change of most conversion anodes during cycling. Herein, a nitrogen-doping graphene/Fe2O3 (N-GF-300) composite material was successfully prepared as a sodium-ion storage anode for sodium ion batteries and sodium ion supercapacitors through a water bath and an annealing process, where Fe2O3 nanoparticles with a homogenous size of about 30 nm were uniformly anchored on the graphene nanosheets. The nitrogen-doping graphene structure enhanced the connection between Fe2O3 nanoparticles with graphene nanosheets to improve electrical conductivity and buffer the volume change of the material for high capacity and stable cycle performance. The N-GF-300 anode material delivered a high reversible discharge capacity of 638 mAh g-1 at a current density of 0.1 A g-1 and retained 428.3 mAh g-1 at 0.5 A g-1 after 100 cycles, indicating a strong cyclability of the SIBs. The asymmetrical N-GF-300//graphene SIC exhibited a high energy density and power density with 58 Wh kg-1 at 1365 W kg-1 in organic solution. The experimental results from this work clearly illustrate that the nitrogen-doping graphene/Fe2O3 composite material N-GF-300 is a potential feasibility for sodium-ion storage devices, which further reveals that the nitrogen doping approach is an effective technique for modifying carbon matrix composites for high reaction kinetics during cycles in sodium-ion storage devices and even other electrochemical storage devices.

Foreground Gating and Background Refining Network for Surveillance Object Detection.

Detecting objects in surveillance videos is an important problem due to its wide applications in traffic control and public security. Existing methods tend to face performance degradation because of false positive or misalignment problems. We propose a novel framework, namely, Foreground Gating and Background Refining Network (FG-BR Net), for surveillance object detection (SOD). To reduce false positives in background regions, which is a critical problem in SOD, we introduce a new module that first subtracts the background of a video sequence and then generates high-quality region proposals. Unlike previous background subtraction methods that may wrongly remove the static foreground objects in a frame, a feedback connection from detection results to background subtraction process is proposed in our model to distill both static and moving objects in surveillance videos. Furthermore, we introduce another module, namely, the background refining stage, to refine the detection results with more accurate localizations. Pairwise non-local operations are adopted to cope with the misalignments between the features of original and background frames. Extensive experiments on real-world traffic surveillance benchmarks demonstrate the competitive performance of the proposed FG-BR Net. In particular, FG-BR Net ranks on the top among all the methods on hard and sunny subsets of the UA-DETRAC detection dataset, without any bells and whistles.

Functional Brain Connectivity Revealed by Sparse Coding of Large-Scale Local Field Potential Dynamics.

Exploration of brain dynamics patterns has attracted increasing attention due to its fundamental significance in understanding the working mechanism of the brain. However, due to the lack of effective modeling methods, how the simultaneously recorded LFP can inform us about the brain dynamics remains a general challenge. In this paper, we propose a novel sparse coding based method to investigate brain dynamics of freely-behaving mice from the perspective of functional connectivity, using super-long local field potential (LFP) recordings from 13 distinct regions of the mouse brain. Compared with surrogate datasets, six and four reproducible common functional connectivities were discovered to represent the space of brain dynamics in the frequency bands of alpha and theta respectively. Modeled by a finite state machine, temporal transition framework of functional connectivities was inferred for each frequency band, and evident preference was discovered. Our results offer a novel perspective for analyzing neural recording data at such high temporal resolution and recording length, as common functional connectivities and their transition framework discovered in this work reveal the nature of the brain dynamics in freely behaving mice.

Identifying Brain Networks at Multiple Time Scales via Deep Recurrent Neural Network.

For decades, task functional magnetic resonance imaging has been a powerful noninvasive tool to explore the organizational architecture of human brain function. Researchers have developed a variety of brain network analysis methods for task fMRI data, including the general linear model, independent component analysis, and sparse representation methods. However, these shallow models are limited in faithful reconstruction and modeling of the hierarchical and temporal structures of brain networks, as demonstrated in more and more studies. Recently, recurrent neural networks (RNNs) exhibit great ability of modeling hierarchical and temporal dependence features in the machine learning field, which might be suitable for task fMRI data modeling. To explore such possible advantages of RNNs for task fMRI data, we propose a novel framework of a deep recurrent neural network (DRNN) to model the functional brain networks from task fMRI data. Experimental results on the motor task fMRI data of Human Connectome Project 900 subjects release demonstrated that the proposed DRNN can not only faithfully reconstruct functional brain networks, but also identify more meaningful brain networks with multiple time scales which are overlooked by traditional shallow models. In general, this work provides an effective and powerful approach to identifying functional brain networks at multiple time scales from task fMRI data.

Ensembles of change-point detectors: implications for real-time BMI applications.

Brain-machine interfaces (BMIs) have been widely used to study basic and translational neuroscience questions. In real-time closed-loop neuroscience experiments, many practical issues arise, such as trial-by-trial variability, and spike sorting noise or multi-unit activity. In this paper, we propose a new framework for change-point detection based on ensembles of independent detectors in the context of BMI application for detecting acute pain signals. Motivated from ensemble learning, our proposed "ensembles of change-point detectors" (ECPDs) integrate multiple decisions from independent detectors, which may be derived based on data recorded from different trials, data recorded from different brain regions, data of different modalities, or models derived from different learning methods. By integrating multiple sources of information, the ECPDs aim to improve detection accuracy (in terms of true positive and true negative rates) and achieve an optimal trade-off of sensitivity and specificity. We validate our method using computer simulations and experimental recordings from freely behaving rats. Our results have shown superior and robust performance of ECPDS in detecting the onset of acute pain signals based on neuronal population spike activity (or combined with local field potentials) recorded from single or multiple brain regions.

Recognizing Brain States Using Deep Sparse Recurrent Neural Network.

Brain activity is a dynamic combination of different sensory responses and thus brain activity/state is continuously changing over time. However, the brain's dynamical functional states recognition at fast time-scales in task fMRI data have been rarely explored. In this paper, we propose a novel 5-layer deep sparse recurrent neural network (DSRNN) model to accurately recognize the brain states across the whole scan session. Specifically, the DSRNN model includes an input layer, one fully-connected layer, two recurrent layers, and a softmax output layer. The proposed framework has been tested on seven task fMRI data sets of Human Connectome Project. Extensive experiment results demonstrate that the proposed DSRNN model can accurately identify the brain's state in different task fMRI data sets and significantly outperforms other auto-correlation methods or non-temporal approaches in the dynamic brain state recognition accuracy. In general, the proposed DSRNN offers a new methodology for basic neuroscience and clinical research.

Large-Scale Circuitry Interactions Upon Earthquake Experiences Revealed by Recurrent Neural Networks.

Brain dynamics has recently received increasing interest due to its significant importance in basic and clinical neurosciences. However, due to inherent difficulties in both data acquisition and data analysis methods, studies on large-scale brain dynamics of mouse with local field potential (LFP) recording are very rare. In this paper, we did a series of works on modeling large-scale mouse brain dynamic activities responding to fearful earthquake. Based on LFP recording data from 13 brain regions that are closely related to fear learning and memory and the effective Bayesian connectivity change point model, we divided the response time series into four stages: "Before," "Earthquake," "Recovery," and "After." We first reported the changes in power and theta-gamma coupling during stage transitions. Then, a recurrent neural network model was designed to model the functional dynamics in these thirteen brain regions and six frequency bands in response to the fear stimulus. Interestingly, our results showed that the functional brain connectivities in theta and gamma bands exhibited distinct response processes: in theta band, there is a separated-united-separated alternation in whole-brain connectivity and a low-high-low change in connectivity strength; however, gamma bands have a united-separated-united transition and a high-low-high alternation in connectivity pattern and strength. In general, our results offer a novel perspective in studying functional brain dynamics under fearful stimulus and reveal its relationship to the brain's structural connectivity substrates.

Expression of HERC4 in lung cancer and its correlation with clinicopathological parameters.

BACKGROUND: Growing evidence suggests that the members of the ubiquitin-proteasome system (UPS) are important for tumorigenesis. HERC4, one component, is a recently identified ubiqutin ligase. However, the expression level and function role of HERC4 in lung cancer remain unknown. Our objective was to investigate any correlation between HERC4 and development of lung cancer and its clinical significance. MATERIALS AND METHODS: To determine HERC4 expression in lung cancer, an immunohistochemistry analysis of a tissue microarray containing samples of 10 lung normal tissues, 15 pulmonary neuroendocrine carcinomas, 45 squamous epithelial cancers and 50 adenocarcinomas was conducted. Receiver operating characteristic (ROC) curve analysis was applied to obtain a cut-off point of 52.5%, above which the expression of HERC4 was regarded as "positive". RESULTS: On the basis of ROC curve analysis, positive expression of HERC4 was detected in 0/10 (0.0%) of lung normal tissues, in 4/15 (26.7%) of pulmonary neuroendocrine carcinomas, in 13/45 (28.9%) of squamous epithelial cancers and in 19/50 (38.0%) of adenocarcinomas. It showed that lung tumors expressed more HERC4 protein than adjacent normal tissues (χ2=4.675, p=0.031). Furthermore, HERC4 positive expression had positive correlation with pT status (χ2=44.894, p=0.000), pN status (χ2=43.628, p=0.000), histological grade (χ2=7.083, p=0.029) and clinical stage (χ2=72.484, p=0.000), but not age (χ2=0.910, p=0.340). CONCLUSIONS: Our analysis suggested that HERC4 is likely to be a diagnostic biomarker for lung cancer.

Characterizing and Differentiating Brain State Dynamics via Hidden Markov Models.

Functional connectivity measured from resting state fMRI (R-fMRI) data has been widely used to examine the brain's functional activities and has been recently used to characterize and differentiate brain conditions. However, the dynamical transition patterns of the brain's functional states have been less explored. In this work, we propose a novel computational framework to quantitatively characterize the brain state dynamics via hidden Markov models (HMMs) learned from the observations of temporally dynamic functional connectomics, denoted as functional connectome states. The framework has been applied to the R-fMRI dataset including 44 post-traumatic stress disorder (PTSD) patients and 51 normal control (NC) subjects. Experimental results show that both PTSD and NC brains were undergoing remarkable changes in resting state and mainly transiting amongst a few brain states. Interestingly, further prediction with the best-matched HMM demonstrates that PTSD would enter into, but could not disengage from, a negative mood state. Importantly, 84% of PTSD patients and 86% of NC subjects are successfully classified via multiple HMMs using majority voting.

Atomic connectomics signatures for characterization and differentiation of mild cognitive impairment.

In recent years, functional connectomics signatures have been shown to be a very valuable tool in characterizing and differentiating brain disorders from normal controls. However, if the functional connectivity alterations in a brain disease are localized within sub-networks of a connectome, then accurate identification of such disease-specific sub-networks is critical and this capability entails both fine-granularity definition of connectome nodes and effective clustering of connectome nodes into disease-specific and non-disease-specific sub-networks. In this work, we adopted the recently developed DICCCOL (dense individualized and common connectivity-based cortical landmarks) system as a fine-granularity high-resolution connectome construction method to deal with the first issue, and employed an effective variant of non-negative matrix factorization (NMF) method to pinpoint disease-specific sub-networks, which we called atomic connectomics signatures in this work. We have implemented and applied this novel framework to two mild cognitive impairment (MCI) datasets from two different research centers, and our experimental results demonstrated that the derived atomic connectomics signatures can effectively characterize and differentiate MCI patients from their normal controls. In general, our work contributed a novel computational framework for deriving descriptive and distinctive atomic connectomics signatures in brain disorders.

Circumcision plus antibiotic, anti-inflammatory, and α-blocker therapy for the treatment for chronic prostatitis/chronic pelvic pain syndrome: a prospective, randomized, multicenter trial.

PURPOSE: The purpose of the study was to evaluate the efficacy of circumcision combined with antibiotic, anti-inflammatory, and α-blocker therapy for the treatment for chronic prostatitis/chronic pelvic pain syndrome (CP/CPPS). METHODS: Subjects assigned to the circumcision group were given antibiotic, anti-inflammatory, and α-blocker medications and scheduled for surgery the same period in each site by study clinicians. Subjects assigned to the control group were asked to only take the same medications and remain uncircumcised until the end of the 3-month study period. The primary outcome was a reduction of at least four points on the National Institutes of Health Chronic Prostatitis Symptom Index (NIH-CPSI). RESULTS: A total of 774 eligible participants underwent randomization, and the ratio of men with a decrease of at least four points on the total NIH-CPSI score from baseline to 12 weeks was 84.6% in the circumcision group and 68.5% in the control group (P < 0.001). Of the 713 men who completed the trial, the median total NIH-CPSI score decreased significantly from 21.0 ± 7.0 to 12.0 ± 8.0 (P < 0.001) in the circumcision group, and in the control group, the change was from 21.0 ± 8.0 to 15.0 ± 7.0 (P < 0.001). Comparison of the changes in the total and three subdomain NIH-CPSI scores over time revealed significant differences between the circumcision and control groups (P < 0.001). CONCLUSIONS: Our findings show that circumcision plus antibiotic, anti-inflammatory, and α-blocker therapy for CP/CPPS patients resulted in improved NIH-CPSI scores compared with medication therapy only.