GlmS plays a key role in the virulence factor expression and biofilm formation ability of Staphylococcus aureus promoted by advanced glycation end products.

Staphylococcus aureus (S. aureus) is well known for its biofilm formation ability and is responsible for serious, chronic refractory infections worldwide. We previously demonstrated that advanced glycation end products (AGEs), a hallmark of chronic hyperglycaemia in diabetic tissues, enhanced biofilm formation by promoting eDNA release via sigB upregulation in S. aureus, contributing to the high morbidity and mortality of patients presenting a diabetic foot ulcer infection. However, the exact regulatory network has not been completely described. Here, we used pull-down assay and LC-MS/MS to identify the GlmS as a candidate regulator of sigB in S. aureus stimulated by AGEs. Dual-luciferase assays and electrophoretic mobility shift assays (EMSAs) revealed that GlmS directly upregulated the transcriptional activity of sigB. We constructed NCTC 8325 ∆glmS for further validation. qRT-PCR analysis revealed that AGEs promoted both glmS and sigB expression in the NCTC 8325 strain but had no effect on NCTC 8325 ∆glmS. NCTC 8325 ∆glmS showed a significant attenuation in biofilm formation and virulence factor expression, accompanied by a decrease in sigB expression, even under AGE stimulation. All of the changes, including pigment deficiency, decreased haemolysis ability, downregulation of hla and hld expression, and less and sparser biofilms, indicated that sigB and biofilm formation ability no longer responded to AGEs in NCTC 8325 ∆glmS. Our data extend the understanding of GlmS in the global regulatory network of S. aureus and demonstrate a new mechanism by which AGEs can upregulate GlmS, which directly regulates sigB and plays a significant role in mediating biofilm formation and virulence factor expression.

Phloretin prolongs lifespan of Caenorhabditis elegans via inhibition of NDUFS1 and NDUFS6 at mitochondrial complex â .

Phloretin has been widely perceived as an antioxidant. However, the bioavailability of phloretin in vivo is generally far too low to elicit a direct antioxidant effect by scavenging reactive oxygen species (ROS). Here we showed that administration of phloretin of apple polyphenols extended lifespan of Caenorhabditis elegans and promoted fitness. Specially phloretin enhanced the survival rates of nematodes under oxidants in an inverted U-shaped dose-response manner. The lifespan-extending effects of phloretin were mediated by ROS via mitochondrial complex I inhibition. The increase of ROS stimulated p38 MAPK/PMK-1 as well as transcription factors of NRF2/SKN-1 and FOXO/DAF-16. Consistent with the involvement of NRF2/SKN-1 and FOXO/DAF-16 in lifespan-extending effects, activities of superoxide dismutase (SOD) and catalase (CAT) were enhanced by phloretin. The exogenous application of antioxidants butylated hydroxyanisole and N-acetylcysteine abolished the increase of ROS, the enhancement of SOD and CAT activities, and the lifespan extending effects of phloretin. Meanwhile, with the inhibition of mitochondrial complex I, ATP was instantly decreased. Both energy sensors of AMPK/AAK-2 and SIRT1/SIR-2.1 were involved in the lifespan extension by phloretin. Transcriptomic, real-time qPCR and molecular docking analyses demonstrated that the binding of phloretin at complex I located at NDUFS1/NUO-5, NDUFS2/GAS-1, and NDUFS6/NDUF-6. The molecular dynamic simulation and binding free energy calculations showed that phloretin had high binding affinities towards NDUFS1 (-7.21 kcal/mol) and NDUFS6 (-7.02 kcal/mol). Collectively, our findings suggested phloretin had effects of life expectancy enhancement and fitness promotion via redox regulations in vivo. NDUFS1/NUO-5 and NDUFS6/NDUF-6 might be new targets in the lifespan and wellness regulations.

Causal associations of genetically predicted gut microbiota and blood metabolites with inflammatory states and risk of infections: a Mendelian randomization analysis.

Background: Inflammation serves as a key pathologic mediator in the progression of infections and various diseases, involving significant alterations in the gut microbiome and metabolism. This study aims to probe into the potential causal relationships between gut microbial taxa and human blood metabolites with various serum inflammatory markers (CRP, SAA1, IL-6, TNF-α, WBC, and GlycA) and the risks of seven common infections (gastrointestinal infections, dysentery, pneumonia, bacterial pneumonia, bronchopneumonia and lung abscess, pneumococcal pneumonia, and urinary tract infections). Methods: Two-sample Mendelian randomization (MR) analysis was performed using inverse variance weighted (IVW), maximum likelihood, MR-Egger, weighted median, and MR-PRESSO. Results: After adding other MR models and sensitivity analyses, genus Roseburia was simultaneously associated adversely with CRP (Beta IVW = -0.040) and SAA1 (Beta IVW = -0.280), and family Bifidobacteriaceae was negatively associated with both CRP (Beta IVW = -0.034) and pneumonia risk (Beta IVW = -0.391). After correction by FDR, only glutaroyl carnitine remained significantly associated with elevated CRP levels (Beta IVW = 0.112). Additionally, threonine (Beta IVW = 0.200) and 1-heptadecanoylglycerophosphocholine (Beta IVW = -0.246) were found to be significantly associated with WBC levels. Three metabolites showed similar causal effects on different inflammatory markers or infectious phenotypes, stearidonate (18:4n3) was negatively related to SAA1 and urinary tract infections, and 5-oxoproline contributed to elevated IL-6 and SAA1 levels. In addition, 7-methylguanine showed a positive correlation with dysentery and bacterial pneumonia. Conclusion: This study provides novel evidence confirming the causal effects of the gut microbiome and the plasma metabolite profile on inflammation and the risk of infection. These potential molecular alterations may aid in the development of new targets for the intervention and management of disorders associated with inflammation and infections.

Directed differentiation of human embryonic stem cells into parathyroid cells and establishment of parathyroid organoids.

Differentiation of human embryonic stem cells (hESCs) into human embryonic stem cells-derived parathyroid-like cells (hESC-PT) has clinical significance in providing new therapies for congenital and acquired parathyroid insufficiency conditions. However, a highly reproducible, well-documented method for parathyroid differentiation remains unavailable. By imitating the natural process of parathyroid embryonic development, we proposed a new hypothesis about the in vitro differentiation of parathyroid-like cells. Transcriptome, differentiation marker protein detection and parathyroid hormone (PTH) secretion assays were performed after the completion of differentiation. To optimize the differentiation protocol and further improve the differentiation rate, we designed glial cells missing transcription factor 2 (GCM2) overexpression lentivirus transfection assays and constructed hESCs-derived parathyroid organoids. The new protocol enabled hESCs to differentiate into hESC-PT. HESC-PT cells expressed PTH, GCM2 and CaSR proteins, low extracellular calcium culture could stimulate hESC-PT cells to secrete PTH. hESC-PT cells overexpressing GCM2 protein secreted PTH earlier than their counterpart hESC-PT cells. Compared with the two-dimensional cell culture environment, hESCs-derived parathyroid organoids secreted more PTH. Both GCM2 lentiviral transfection and three-dimensional cultures could make hESC-PT cells functionally close to human parathyroid cells. Our study demonstrated that hESCs could differentiate into hESC-PT in vitro, which paves the road for applying the technology to treat hypoparathyroidism and introduces new approaches in the field of regenerative medicine.

sEMG-Based Inter-Session Hand Gesture Recognition via Domain Adaptation with Locality Preserving and Maximum Margin.

Surface electromyography (sEMG)-based gesture recognition can achieve high intra-session performance. However, the inter-session performance of gesture recognition decreases sharply due to the shift in data distribution. Therefore, developing a robust model to minimize the data distribution difference is crucial to improving the user experience. In this work, based on the inter-session gesture recognition task, we propose a novel algorithm called locality preserving and maximum margin criterion (LPMM). The LPMM algorithm integrates three main modules, including domain alignment, pseudo-label selection, and iteration result selection. Domain alignment is designed to preserve the neighborhood structure of the feature and minimize the overlap of different classes. The pseudo-label selection and iteration result selection can avoid the decrease in accuracy caused by mislabeled samples. The proposed algorithm was evaluated on two of the most widely used EMG databases. It achieves a mean accuracy of 98.46% and 71.64%, respectively, which is superior to state-of-the-art domain adaptation methods.

A Sequential End-to-End Neonatal Sleep Staging Model with Squeeze and Excitation Blocks and Sequential Multi-Scale Convolution Neural Networks.

Automatic sleep staging offers a quick and objective assessment for quantitatively interpreting sleep stages in neonates. However, most of the existing studies either do not encompass any temporal information, or simply apply neural networks to exploit temporal information at the expense of high computational overhead and modeling ambiguity. This limits the application of these methods to multiple scenarios. In this paper, a sequential end-to-end sleep staging model, SeqEESleepNet, which is competent for parallelly processing sequential epochs and has a fast training rate to adapt to different scenarios, is proposed. SeqEESleepNet consists of a sequence epoch generation (SEG) module, a sequential multi-scale convolution neural network (SMSCNN) and squeeze and excitation (SE) blocks. The SEG module expands independent epochs into sequential signals, enabling the model to learn the temporal information between sleep stages. SMSCNN is a multi-scale convolution neural network that can extract both multi-scale features and temporal information from the signal. Subsequently, the followed SE block can reassign the weights of features through mapping and pooling. Experimental results exhibit that in a clinical dataset, the proposed method outperforms the state-of-the-art approaches, achieving an overall accuracy, F1-score, and Kappa coefficient of 71.8%, 71.8%, and 0.684 on a three-class classification task with a single channel EEG signal. Based on our overall results, we believe the proposed method could pave the way for convenient multi-scenario neonatal sleep staging methods.

Polygenic risk score predicts all-cause death in East Asian patients with prior coronary artery disease.

Introduction: Coronary artery disease (CAD) is a highly heritable and multifactorial disease. Numerous genome-wide association studies (GWAS) facilitated the construction of polygenic risk scores (PRS) for predicting future incidence of CAD, however, exclusively in European populations. Furthermore, identifying CAD patients with elevated risks of all-cause death presents a critical challenge in secondary prevention, which will contribute largely to reducing the burden for public healthcare. Methods: We recruited a cohort of 1,776 Chinese CAD patients and performed medical follow-up for up to 11 years. A pruning and thresholding method was used to calculate PRS of CAD and its 14 risk factors. Their correlations with all-cause death were computed via Cox regression. Results and discussion: We found that the PRS for CAD and its seven risk factors, namely myocardial infarction, ischemic stroke, angina, heart failure, low-density lipoprotein cholesterol, total cholesterol and C-reaction protein, were significantly associated with death (P ≤ 0.05), whereas the PRS of body mass index displayed moderate association (P < 0.1). Elastic-net Cox regression with 5-fold cross-validation was used to integrate these nine PRS models into a meta score, metaPRS, which performed well in stratifying patients at different risks for death (P < 0.0001). Combining metaPRS with clinical risk factors further increased the discerning power and a 4% increase in sensitivity. The metaPRS generated from the genetic susceptibility to CAD and its risk factors can well stratify CAD patients by their risks of death. Integrating metaPRS and clinical risk factors may contribute to identifying patients at higher risk of poor prognosis.

A Novel Ultrafiltrate Extract of Propolis Exerts Anti-inflammatory Activity through Metabolic Rewiring.

Thousands of years ago, humans started to use propolis because of its medicinal properties, and modern science has successfully identified several bioactive molecules within this resinous bee product. However, a natural propolis extract which has been removed the adhesive glue and preserved propolis bioactive compounds is urgently needed to maximise the therapeutic opportunities. In this study, a novel ultrafiltrate fraction from Brazilian green propolis, termed P30K, was demonstrated with anti-inflammatory properties, both in vitro and in vivo. Total flavonoids and total phenolic acids content in P30K were 244.6 mg/g and 275.8 mg/g respectively, while the IC50 value of inhibition of cyclooxygenase-2 (COX-2) was 8.30 µg/mL. The anti-inflammatory activity of P30K was furtherly corroborated in experimental models of lipopolysaccharides (LPS)-induced acute liver and lung injury. Mechanistically, integrated GC-MS and LC-MS based serum metabolomics analysis revealed that P30K modulated citrate cycle (TCA), pyruvate, glyoxylate and dicarboxylate metabolism pathways to inhibit secretion of pro-inflammatory cytokines. Results of network pharmacology and molecular docking suggested that P30K targeted catechol-O-methyltransferases (COMT), 11ß-hydroxysteroid dehydrogenases (HSD11B1), and monoamine oxidases (MAOA and MAOB) to promote cellular metabolomic rewiring. Collectively, our work reveals P30K as an efficient therapeutic agent against inflammatory conditions and its efficacy is related to metabolic rewiring.

Towards Real-Time Sleep Stage Prediction and Online Calibration Based on Architecturally Switchable Deep Learning Models.

Despite the recent advances in automatic sleep staging, few studies have focused on real-time sleep staging to promote the regulation of sleep or the intervention of sleep disorders. In this paper, a novel network named SwSleepNet, that can handle both precisely offline sleep staging, and online sleep stages prediction and calibration is proposed. For offline analysis, the proposed network coordinates sequence broadening module (SBM), sequential CNN (SCNN), squeeze and excitation (SE) block, and sequence consolidation module (SCM) to balance the operational efficiency of the network and the comprehensive feature extraction. For online analysis, only SCNN and SE are involved in predicting the sleep stage within a short-time segment of the recordings. Once more than two successive segments have disparate predictions, the calibration mechanism will be triggered, and contextual information will be involved. In addition, to investigate the appropriate time of the segment that is suitable to predict a sleep stage, segments with five-second, three-second, and two-second data are analyzed. The performance of SwSleepNet is validated on two publicly available datasets Sleep-EDF Expanded and Montreal Archive of Sleep Studies (MASS), and one clinical dataset Huashan Hospital Fudan University (HSFU), with the offline accuracy of 84.5%, 86.7%, and 81.8%, respectively, which outperforms the state-of-the-art methods. Additionally, for the online sleep staging, the dedicated calibration mechanism allows SwSleepNet to achieve high accuracy over 80% on three datasets with the short-time segments, demonstrating the robustness and stability of SwSleepNet. This study presents a real-time sleep staging architecture, which is expected to pave the way for accurate sleep regulation and intervention.

Membrane-Based Preparation Process and Antioxidant and Anti-AGEs Activities of a Novel Propolis Ultrafiltrate.

Propolis is one functional supplement with hundreds of years of usage. However, it's rarely consumed directly for its resinous property. Herein, a pre-treated process which can remove the impurity while preserve its bioactivities is needed to maximise its therapeutic opportunities. In the present study, a membrane-based ultrafiltration process was developed on a KM1812-NF experimental instrument. Using Brazilian green propolis as testing material, all experimental steps and parameters were sequentially optimized. In addition, a mathematical model was developed to fit the process. As a result, the optimum solvent was 60 % ethanol adjusted to pH 8-9, while the optimum MWCO (molecular weight cut-off) value of membrane was 30 KDa. The membrane filtration dynamic model fitted with the function y=(ax+b)/(1+cx+dx2 ). The resulting propolis ultrafiltrate from Brazilian green propolis, termed P30K, contains the similar profile of flavonoids and phenolic acids as raw propolis. Meanwhile, the ORAC (oxygen radical absorbance capacity) value of P30K is 11429.45±1557.58 µM TE/g and the IC50 value of inhibition of fluorescent AGEs (advanced glycation end products) formation is 0.064 mg/mL. Our work provides an innovative alternative process for extraction of active compounds from propolis and reveals P30K as an efficient therapeutic antioxidant.

Physiological measurements for driving drowsiness: A comparative study of multi-modality feature fusion and selection.

A large number of traffic accidents were caused by drowsiness while driving. In-vehicle alert system based on physiological signals was one of the most promising solutions to monitor driving fatigue. However, different physiological modalities can be used, and many relative studies compared different modalities without considering the implementation feasibility of portable or wearable devices. Moreover, evaluations of each modality in previous studies were based on inconsistent choices of fatigue label and signal features, making it hard to compare the results of different studies. Therefore, the modality comparison and fusion for continuous drowsiness estimation while driving was still unclear. This work sought to comprehensively compare widely-used physiological modalities, including forehead electroencephalogram (EEG), electrooculogram (EOG), R-R intervals (RRI) and breath, in a hardware setting feasible for portable or wearable devices to monitor driving fatigue. Moreover, a more general conclusion on modality comparison and fusion was reached based on the regression of features or their combinations and the awake-to-drowsy transition. Finally, the feature subset of fused modalities was produced by feature selection method, to select the optimal feature combination and reduce computation consumption. Considering practical feasibility, the most effective combination with the highest correlation coefficient was using forehead EEG or EOG, along with RRI and RRI-derived breath. If more comfort and convenience was required, the combination of RRI and RRI-derived breath was also promising.

MKL-1 suppresses ferroptosis by activating system Xc- and increasing glutathione synthesis.

Chemotherapy is a standard method in traditional treatment for gastric cancer. It is well known that the anti-tumor effects of chemotherapy are achieved mainly through the direct killing of cancer cells via apoptosis. However, chemotherapy often fails due to drug resistance. Therefore, non-apoptotic cell death induction by ferroptosis has recently been proposed as a new therapeutic modality to ablate cancer. In this study, we determined the role of MKL-1 in ferroptosis. In vitro and in vivo experiments showed that inhibition of MKL-1 expression significantly enhanced cell sensitivity to ferroptosis-inducing agents. It functions by targeting system Xc- to affect the synthesis of GSH in cells. Therefore, we developed an exosome-based therapeutic approach targeting MKL-1, which provides a novel insight into the treatment of gastric cancer.

Research status of soda residue in the field of environmental pollution control.

High-quality soda ash (Na2CO3) is mainly produced using the ammonia-alkaline method, generating a significant amount of industrial waste called soda residue. In China, the annual production of soda residue exceeds 10 million tons. The large-scale open-air storage of soda residue not only occupies land but also causes severe pollution to the surrounding environment. Soda residue displays characteristics such as strong alkalinity, high reactivity, and a well-developed pore structure, making it a valuable raw material for producing environmentally functional materials. This article provided an overview and summary of soda residue, including its sources and hazards, basic properties, applications in environmental management (wastewater treatment, flue gas desulfurization, and soil remediation), and associated risks. The limitations of using soda residue in "waste to waste" technologies were also analyzed. Based on this analysis, the article suggests focusing on simultaneous removal of heavy metal ions using soda residue, safely disposing of and acquiring resources from metal-laden sludge, efficiently dechlorinating soda residue, using soda residue for contaminated soil solidification, stabilization, and assisted remediation, controlling pollution via green and circular utilization approaches, and assessing long-term risk.

SDE-YOLO: A Novel Method for Blood Cell Detection.

This paper proposes an improved target detection algorithm, SDE-YOLO, based on the YOLOv5s framework, to address the low detection accuracy, misdetection, and leakage in blood cell detection caused by existing single-stage and two-stage detection algorithms. Initially, the Swin Transformer is integrated into the back-end of the backbone to extract the features in a better way. Then, the 32 × 32 network layer in the path-aggregation network (PANet) is removed to decrease the number of parameters in the network while increasing its accuracy in detecting small targets. Moreover, PANet substitutes traditional convolution with depth-separable convolution to accurately recognize small targets while maintaining a fast speed. Finally, replacing the complete intersection over union (CIOU) loss function with the Euclidean intersection over union (EIOU) loss function can help address the imbalance of positive and negative samples and speed up the convergence rate. The SDE-YOLO algorithm achieves a mAP of 99.5%, 95.3%, and 93.3% on the BCCD blood cell dataset for white blood cells, red blood cells, and platelets, respectively, which is an improvement over other single-stage and two-stage algorithms such as SSD, YOLOv4, and YOLOv5s. The experiment yields excellent results, and the algorithm detects blood cells very well. The SDE-YOLO algorithm also has advantages in accuracy and real-time blood cell detection performance compared to the YOLOv7 and YOLOv8 technologies.

Removal of hydrogen sulfide in the gas phase by carbide slag modified bentonite.

Bentonite-based adsorbents for the removal of hydrogen sulfide (H2S) were prepared by a wet-mixing method using carbide slag as the active component. The effects of carbide slag content, calcination temperature, calcination time, and reaction temperature on the H2S adsorption capacity were investigated. The results showed that compared with the blank bentonite adsorbent, the carbide slag-modified bentonite-based adsorbent enhanced the chemisorption of H2S. The adsorption capacity of the carbide slag modified bentonite adsorbent (2.50 mg g-1) was more than 40 times higher than that of the blank bentonite-based adsorbent (0.06 mg g-1) under optimal conditions. The optimal conditions for H2S removal were 3 : 5 ratio of carbide slag-to-bentonite, calcination temperature of 450 °C for 2 h, and reaction temperature of 95 °C. H2S was mainly removed in the mesopores and macropores of the adsorbent and was finally transformed to CaS and sulfate on the adsorbent surface. The adsorption process of H2S followed the Freundlich adsorption isotherm equation and Bangham adsorption kinetic model.

The regulatory effects of fucoidan and laminarin on functional dyspepsia mice induced by loperamide.

Gastrointestinal dysmotility is a common cause of functional dyspepsia. As two kinds of polysaccharides derived from brown algae, fucoidan and laminarin possess many physiological properties; however, their relative abilities in regulating gastrointestinal motility have not been illustrated yet. In this study, we aimed to investigate the regulatory effect of fucoidan and laminarin on functional dyspepsia mice induced by loperamide. Mice with gastrointestinal dysmotility were treated with fucoidan (100 and 200 mg per kg bw) and laminarin (50 and 100 mg per kg bw). As a result, fucoidan and laminarin reversed the dysfunction mainly through regulating gastrointestinal hormones (motilin and ghrelin), the cholinergic pathway, the total bile acid level, c-kit protein expression, and gastric smooth muscle contraction-related gene expression (ANO1 and RYR3). Moreover, fucoidan and laminarin intervention modulated the gut microbiota profile including the altered richness of Muribaculaceae, Lachnospiraceae, and Streptococcus. The results indicated that fucoidan and laminarin may restore the rhythm of the migrating motor complex and regulate gut microecology. In conclusion, we provided evidence to support that fucoidan and laminarin might have potential abilities to regulate gastrointestinal motility.

A consortium of three-bacteria isolated from human feces inhibits formation of atherosclerotic deposits and lowers lipid levels in a mouse model.

By a survey of metagenome-wide association studies (MWAS), we found a robust depletion of Bacteroides cellulosilyticus, Faecalibacterium prausnitzii, and Roseburia intestinalis in individuals with atherosclerotic cardiovascular disease (ACVD). From an established collection of bacteria isolated from healthy Chinese individuals, we selected B. cellulosilyticus, R. intestinalis, and Faecalibacterium longum, a bacterium related to F. prausnitzii, and tested the effects of these bacteria in an Apoe/- atherosclerosis mouse model. We show that administration of these three bacterial species to Apoe-/- mice robustly improves cardiac function, reduces plasma lipid levels, and attenuates the formation of atherosclerotic plaques. Comprehensive analysis of gut microbiota, plasma metabolome, and liver transcriptome revealed that the beneficial effects are associated with a modulation of the gut microbiota linked to a 7α-dehydroxylation-lithocholic acid (LCA)-farnesoid X receptor (FXR) pathway. Our study provides insights into transcriptional and metabolic impact whereby specific bacteria may hold promises for prevention/treatment of ACVD.

Targeting the inward rectifier potassium channel 5.1 in thyroid cancer: artificial intelligence-facilitated molecular docking for drug discovery.

BACKGROUND: Recurrent and metastatic thyroid cancer is more invasive and can transform to dedifferentiated thyroid cancer, thus leading to a severe decline in the 10-year survival. The thyroid-stimulating hormone receptor (TSHR) plays an important role in differentiation process. We aim to find a therapeutic target in redifferentiation strategies for thyroid cancer. METHODS: Our study integrated the differentially expressed genes acquired from the Gene Expression Omnibus database by comparing TSHR expression levels in the Cancer Genome Atlas database. We conducted functional enrichment analysis and verified the expression of these genes by RT-PCR in 68 pairs of thyroid tumor and paratumor tissues. Artificial intelligence-enabled virtual screening was combined with the VirtualFlow platform for deep docking. RESULTS: We identified five genes (KCNJ16, SLC26A4, TG, TPO, and SYT1) as potential cancer treatment targets. TSHR and KCNJ16 were downregulated in the thyroid tumor tissues, compared with paired normal tissues. In addition, KCNJ16 was lower in the vascular/capsular invasion group. Enrichment analyses revealed that KCNJ16 may play a significant role in cell growth and differentiation. The inward rectifier potassium channel 5.1 (Kir5.1, encoded by KCNJ16) emerged as an interesting target in thyroid cancer. Artificial intelligence-facilitated molecular docking identified Z2087256678_2, Z2211139111_1, Z2211139111_2, and PV-000592319198_1 (-7.3 kcal/mol) as the most potent commercially available molecular targeting Kir5.1. CONCLUSION: This study may provide greater insights into the differentiation features associated with TSHR expression in thyroid cancer, and Kir5.1 may be a potential therapeutic target in the redifferentiation strategies for recurrent and metastatic thyroid cancer.

MaskSleepNet: A Cross-Modality Adaptation Neural Network for Heterogeneous Signals Processing in Sleep Staging.

Deep learning methods have become an important tool for automatic sleep staging in recent years. However, most of the existing deep learning-based approaches are sharply constrained by the input modalities, where any insertion, substitution, and deletion of input modalities would directly lead to the unusable of the model or a deterioration in the performance. To solve the modality heterogeneity problems, a novel network architecture named MaskSleepNet is proposed. It consists of a masking module, a multi-scale convolutional neural network (MSCNN), a squeezing and excitation (SE) block, and a multi-headed attention (MHA) module. The masking module consists of a modality adaptation paradigm that can cooperate with modality discrepancy. The MSCNN extracts features from multiple scales and specially designs the size of the feature concatenation layer to prevent invalid or redundant features from zero-setting channels. The SE block further optimizes the weights of the features to optimize the network learning efficiency. The MHA module outputs the prediction results by learning the temporal information between the sleeping features. The performance of the proposed model was validated on two publicly available datasets, Sleep-EDF Expanded (Sleep-EDFX) and Montreal Archive of Sleep Studies (MASS), and a clinical dataset, Huashan Hospital Fudan University (HSFU). The proposed MaskSleepNet can achieve favorable performance with input modality discrepancy, e.g. for single-channel EEG signal, it can reach 83.8%, 83.4%, 80.5%, for two-channel EEG+EOG signals it can reach 85.0%, 84.9%, 81.9% and for three-channel EEG+EOG+EMG signals, it can reach 85.7%, 87.5%, 81.1% on Sleep-EDFX, MASS, and HSFU, respectively. In contrast the accuracy of the state-of-the-art approach which fluctuated widely between 69.0% and 89.4%. The experimental results exhibit that the proposed model can maintain superior performance and robustness in handling input modality discrepancy issues.

MS-HNN: Multi-Scale Hierarchical Neural Network With Squeeze and Excitation Block for Neonatal Sleep Staging Using a Single-Channel EEG.

Most existing neonatal sleep staging appro- aches applied multiple EEG channels to obtain good performance. However, it potentially increased the computational complexity and led to an increased risk of skin disruption to neonates during data acquisition. In this paper, a multi-scale hierarchical neural network (MS-HNN) with a squeeze and excitation (SE) block for neonatal sleep staging is presented in this study on the basis of a single EEG channel. MS-HNN composes of multi-scale convolutional neural network (MSCNN), temporal information learning (TIL) module, and squeeze and excitation (SE) block. MSCNN can extract features from different scales and frequencies, and TIL module can acquire the transition information among adjacent stages. In addition, for these extracted features, SE block can selectively concentrate on informative features and weaken redundant features for achieving better performance. The proposed approach was validated on a clinical dataset involving 64 neonates from the Children's Hospital of Fudan University (CHFU). The proposed network achieves an accuracy of 75.4% and 76.5% for three-class automatic neonatal sleep staging with the single-EEG channel and the eight-EEG channels, respectively. The experimental results show that the proposed method can maintain good performance by making full use of the information in the single channel while reducing the channels to control the computational overhead.