Harnessing the power of natural alkaloids: the emergent role in epilepsy therapy.

The quest for effective epilepsy treatments has spotlighted natural alkaloids due to their broad neuropharmacological effects. This review provides a comprehensive analysis of the antiseizure properties of various natural compounds, with an emphasis on their mechanisms of action and potential therapeutic benefits. Our findings reveal that bioactive substances such as indole, quinoline, terpenoid, and pyridine alkaloids confer medicinal benefits by modulating synaptic interactions, restoring neuronal balance, and mitigating neuroinflammation-key factors in managing epileptic seizures. Notably, these compounds enhance GABAergic neurotransmission, diminish excitatory glutamatergic activities, particularly at NMDA receptors, and suppress proinflammatory pathways. A significant focus is placed on the strategic use of nanoparticle delivery systems to improve the solubility, stability, and bioavailability of these alkaloids, which helps overcome the challenges associated with crossing the blood-brain barrier (BBB). The review concludes with a prospective outlook on integrating these bioactive substances into epilepsy treatment regimes, advocating for extensive research to confirm their efficacy and safety. Advancing the bioavailability of alkaloids and rigorously assessing their toxicological profiles are essential to fully leverage the therapeutic potential of these compounds in clinical settings.

Endothelial HIF2α suppresses retinal angiogenesis in neonatal mice by upregulating NOTCH signaling.

Prolyl hydroxylase domain (PHD) proteins are oxygen sensors that use intracellular oxygen as a substrate to hydroxylate hypoxia-inducible factor (HIF) α proteins, routing them for polyubiquitylation and proteasomal degradation. Typically, HIFα accumulation in hypoxic or PHD-deficient tissues leads to upregulated angiogenesis. Here, we report unexpected retinal phenotypes associated with endothelial cell (EC)-specific gene targeting of Phd2 (Egln1) and Hif2alpha (Epas1). EC-specific Phd2 disruption suppressed retinal angiogenesis, despite HIFα accumulation and VEGFA upregulation. Suppressed retinal angiogenesis was observed both in development and in the oxygen-induced retinopathy (OIR) model. On the other hand, EC-specific deletion of Hif1alpha (Hif1a), Hif2alpha, or both did not affect retinal vascular morphogenesis. Strikingly, retinal angiogenesis appeared normal in mice double-deficient for endothelial PHD2 and HIF2α. In PHD2-deficient retinal vasculature, delta-like 4 (DLL4, a NOTCH ligand) and HEY2 (a NOTCH target) were upregulated by HIF2α-dependent mechanisms. Inhibition of NOTCH signaling by a chemical inhibitor or DLL4 antibody partially rescued retinal angiogenesis. Taken together, our data demonstrate that HIF2α accumulation in retinal ECs inhibits rather than stimulates retinal angiogenesis, in part by upregulating DLL4 expression and NOTCH signaling.

Epilepsy EEG signals classification based on sparse principal component logistic regression model.

Epilepsy is a chronic brain disease caused by excessive discharge of brain neurons. Long-term recurrent seizures bring a lot of trouble to patients and their families. Prediction of different stages of epilepsy is of great significance. We extract pearson correlation coefficients (PCC) between channels in different frequency bands as features of EEG signals for epilepsy stages prediction. However, the features are of large feature dimension and serious multi-collinearity. To eliminate these adverse influence, the combination of traditional dimension reduction method principal component analysis (PCA) and logistic regression method with regularization term is proposed to avoid over-fitting and achieve the feature sparsity. The experiments are conducted on the widely used CHB-MIT dataset using different regularization terms L1 and L2, respectively. The proposed method identifies various stages of epilepsy quickly and efficiently, and it presents the best average accuracy of 94.86%, average precision of 96.71%, average recall of 93.48%, average kappa value of 0.90 and average Matthews correlation coefficient (MCC) value of 0.90 for all patients.

Comparison of the quality of Nurse-Led palliative care with standard medical care during six months in 405 patients with Parkinson's disease and burdens of their Caregivers: A retrospective study at a single center in China.

BACKGROUND: Palliative care is mainly used to improve the quality of life of patients with chronic diseases by addressing their medical conditions and psychological problems. End-stage Parkinson's disease (PD) is also a progressive disease like cancer and could be managed by palliative care. This study was conducted at a single center in China and aimed to compare the quality of nurse-led palliative care with standard medical care during six months in 405 patients with Parkinson's disease (PPD) and their caregivers using the Chinese version of the 39-item Parkinson's Disease Questionnaire (PDQ-39) and the Chinese Zarit Burden Interview (ZBI) scale. METHODS: PPD (stage 2-5) received nurse-led palliative care (NP cohort, 103 patients; 103 caregivers) or neurologist-led standard care (NS cohort, 134 patients; 134 caregivers), or primary care practitioner-led usual care (PS cohort, 168 patients; 168 caregivers) for six months. RESULTS: Before the health professional-led care (BN), the PDQ-39 score of PPD was 68 (71-64) and their caregivers had 54.86 ± 7.64 a ZBI scale. After 6-months of the health professional-led care (AN), the PDQ-39 score of PPD and a ZBI scale of their caregivers decreased for the NP cohort as compared to those of BN condition and those of patients in the NS and PS cohorts at AN condition (p < 0.001 for all). CONCLUSIONS: The quality of life of PPD must be improved and the burden on their caregivers must be relieved. Nurse-led palliative care successfully improved the quality of life of PPD and reduced their caregiver burden.

Diagnosis of Esophageal Squamous Cell Carcinoma by High-Performance Serum Metabolic Fingerprints: A Retrospective Study.

Esophageal squamous cell carcinoma (ESCC) is a highly prevalent and aggressive malignancy, and timely diagnosis of ESCC contributes to an increased cancer survival rate. However, current detection methods for ESCC mainly rely on endoscopic examination, limited by a relatively low participation rate. Herein, ferric-particle-enhanced laser desorption/ionization mass spectrometry (FPELDI MS) is utilized to record the serum metabolic fingerprints (SMFs) from a retrospective cohort (523 non-ESCC participants and 462 ESCC patients) to build diagnostic models toward ESCC. The PFELDI MS achieved high speed (≈30 s per sample), desirable reproducibility (coefficients of variation < 15%), and high throughput (985 samples with ≈124 200 data points for each spectrum). Desirable diagnostic performance with area-under-the-curves (AUCs) of 0.925-0.966 is obtained through machine learning of SMFs. Further, a metabolic biomarker panel is constructed, exhibiting superior diagnostic sensitivity (72.2-79.4%, p < 0.05) as compared with clinical protein biomarker tests (4.3-22.9%). Notably, the biomarker panel afforded an AUC of 0.844 (95% confidence interval [CI]: 0.806-0.880) toward early ESCC diagnosis. This work highlighted the potential of metabolic analysis for accurate screening and early detection of ESCC and offered insights into the metabolic characterization of diseases including but not limited to ESCC.

Dual-Teacher Feature Distillation: A Transfer Learning Method for Insomniac PSG Staging.

Insomnia is the most common sleep disorder linked with adverse long-term medical and psychiatric outcomes. Automatic sleep staging plays a crucial role in aiding doctors to diagnose insomnia disorder. Only a few studies have been conducted to develop automatic sleep staging methods for insomniacs, and most of them have utilized transfer learning methods, which involve pre-training models on healthy individuals and then fine-tuning them on insomniacs. Unfortunately, significant differences in feature distribution between the two subject groups impede the transfer performance, highlighting the need to effectively integrate the features of healthy subjects and insomniacs. In this paper, we propose a dual-teacher cross-domain knowledge transfer method based on the feature-based knowledge distillation to improve the performance of sleep staging for insomniacs. Specifically, the insomnia teacher directly learns from insomniacs and feeds the corresponding domain-specific features into the student network, while the health domain teacher guide the student network to learn domain-generic features. During the training process, we adopt the OFD (Overhaul of Feature Distillation) method to build the health domain teacher. We conducted the experiments to validate the proposed method, using the Sleep-EDF database as the source domain and the CAP-Database as the target domain. The results demonstrate that our method surpasses advanced techniques, achieving an average sleep staging accuracy of 80.56% on the CAP-Database. Furthermore, our method exhibits promising performance on the private dataset.

Simultaneous removal of Cr(â ¥) and tetracycline from wastewater by dielectric barrier discharge plasma coupled with TiO2/rGO/Cu2O composites: Performance and mechanism.

In this study, dielectric barrier discharge (DBD) plasma combined with titanium dioxide/reduced graphene oxide/copper oxide (TiO2/rGO/Cu2O) composites for simultaneous removal of hexavalent chromium (Cr(Ⅵ)) and tetracycline (TC) from wastewater were explored systematically. The TiO2/rGO/Cu2O composites were successfully prepared to improve the specific surface area and charge carrier separation rate. When Cr(Ⅵ) and TC coexist, the two pollutants have better removal efficiency without changing the initial pH. Moreover, the removal efficiency of Cr(Ⅵ) and TC could be further improved in the DBD-TiO2/rGO/Cu2O system, indicating that the TiO2/rGO/Cu2O composites also exhibited good synergistic effects with the DBD plasma. The mechanism exploration showed that the TiO2/rGO/Cu2O composite catalyst could be activated in DBD system to produce various active species by photocatalytic reaction, among which photo-generated electrons and •O2- could significantly enhance Cr(Ⅵ) reduction, while photo-generated holes and •OH could improve TC degradation. More importantly, the intermediate products obtained from TC degradation can be oxidized to •CO2- by photo-generated holes, which can also facilitate the reduction of Cr(Ⅵ). This study shows that DBD combined with TiO2/rGO/Cu2O composites are capable of simultaneous Cr(Ⅵ) reduction and TC degradation, which would provide novel ideas for practical wastewater remediation.

Magnetic resonance imaging radiomics-based prediction of clinically significant prostate cancer in equivocal PI-RADS 3 lesions in the transitional zone.

Purpose: This bi-institutional study aimed to establish a robust model for predicting clinically significant prostate cancer (csPCa) (pathological grade group ≥ 2) in PI-RADS 3 lesions in the transition zone by comparing the performance of combination models. Materials and methods: This study included 243 consecutive men who underwent 3-Tesla magnetic resonance imaging (MRI) and ultrasound-guided transrectal biopsy from January 2020 and April 2022 which is divided into a training cohort of 170 patients and a separate testing cohort of 73 patients. T2WI and DWI images were manually segmented for PI-RADS 3 lesions for the mean ADC and radiomic analysis. Predictive clinical factors were identified using both univariate and multivariate logistic models. The least absolute shrinkage and selection operator (LASSO) regression models were deployed for feature selection and for constructing radiomic signatures. We developed nine models utilizing clinical factors, radiological features, and radiomics, leveraging logistic and XGboost methods. The performances of these models was subsequently compared using Receiver Operating Characteristic (ROC) analysis and the Delong test. Results: Out of the 243 participants with a median age of 70 years, 30 were diagnosed with csPCa, leaving 213 without a csPCa diagnosis. Prostate-specific antigen density (PSAD) stood out as the only significant clinical factor (odds ratio [OR], 1.068; 95% confidence interval [CI], 1.029-1.115), discovered through the univariate and multivariate logistic models. Seven radiomic features correlated with csPCa prediction. Notably, the XGboost model outperformed eight other models (AUC of the training cohort: 0.949, and validation cohort: 0.913). However, it did not surpass the PSAD+MADC model (P > 0.05) in the training and testing cohorts (AUC, 0.949 vs. 0.888 and 0.913 vs. 0.854, respectively). Conclusion: The machine learning XGboost model presented the best performance in predicting csPCa in PI-RADS 3 lesions within the transitional zone. However, the addition of radiomic classifiers did not display any significant enhancement over the compound model of clinical and radiological findings. The most exemplary and generalized option for quantitative prostate evaluation was Mean ADC+PSAD.

MMT: Cross Domain Few-Shot Learning via Meta-Memory Transfer.

Few-shot learning aims to recognize novel categories solely relying on a few labeled samples, with existing few-shot methods primarily focusing on the categories sampled from the same distribution. Nevertheless, this assumption cannot always be ensured, and the actual domain shift problem significantly reduces the performance of few-shot learning. To remedy this problem, we investigate an interesting and challenging cross-domain few-shot learning task, where the training and testing tasks employ different domains. Specifically, we propose a Meta-Memory scheme to bridge the domain gap between source and target domains, leveraging style-memory and content-memory components. The former stores intra-domain style information from source domain instances and provides a richer feature distribution. The latter stores semantic information through exploration of knowledge of different categories. Under the contrastive learning strategy, our model effectively alleviates the cross-domain problem in few-shot learning. Extensive experiments demonstrate that our proposed method achieves state-of-the-art performance on cross-domain few-shot semantic segmentation tasks on the COCO-20 i, PASCAL-5 i, FSS-1000, and SUIM datasets and positively affects few-shot classification tasks on Meta-Dataset.

Sustained Response to Ruxolitinib of Eosinophilia-Associated Myeloproliferative Neoplasm with Translocation t(8;9)(p21;p24).

The translocation t(8;9) produces the fusion gene PCM1-JAK2, resulting in the continuous activation of the JAK2 tyrosine kinase. Myelodysplastic/myeloproliferative neoplasms are the most common disease with t(8;9)/PCM1-JAK2. Individuals with this abnormality have similar features, and JAK2 kinase inhibitor (ruxolitinib) is an effective treatment of the condition. The long-term remission results of ruxolitinib are varied. It is important to determine the response to ruxolitinib. Here, we describe a patient who has been diagnosed with eosinophilia-associated myeloproliferative neoplasm with t(8;9)(p21;p24). This patient has achieved sustained response for >1 year since the administration of ruxolitinib.

Prolonged hypoxia alleviates prolyl hydroxylation-mediated suppression of RIPK1 to promote necroptosis and inflammation.

The prolyl hydroxylation of hypoxia-inducible factor 1α (HIF-1α) mediated by the EGLN-pVHL pathway represents a classic signalling mechanism that mediates cellular adaptation under hypoxia. Here we identify RIPK1, a known regulator of cell death mediated by tumour necrosis factor receptor 1 (TNFR1), as a target of EGLN1-pVHL. Prolyl hydroxylation of RIPK1 mediated by EGLN1 promotes the binding of RIPK1 with pVHL to suppress its activation under normoxic conditions. Prolonged hypoxia promotes the activation of RIPK1 kinase by modulating its proline hydroxylation, independent of the TNFα-TNFR1 pathway. As such, inhibiting proline hydroxylation of RIPK1 promotes RIPK1 activation to trigger cell death and inflammation. Hepatocyte-specific Vhl deficiency promoted RIPK1-dependent apoptosis to mediate liver pathology. Our findings illustrate a key role of the EGLN-pVHL pathway in suppressing RIPK1 activation under normoxic conditions to promote cell survival and a model by which hypoxia promotes RIPK1 activation through modulating its proline hydroxylation to mediate cell death and inflammation in human diseases, independent of TNFR1.

Tailless and hypoxia inducible factor-2α cooperate to sustain proangiogenic states of retinal astrocytes in neonatal mice.

Tailless (TLX, an orphan nuclear receptor) and hypoxia inducible factor-2α (HIF2α) are both essential for retinal astrocyte and vascular development. Tlx-/- mutation and astrocyte specific Hif2α disruption in Hif2αf/f/GFAPCre mice are known to cause defective astrocyte development and block vascular development in neonatal retinas. Here we report that TLX and HIF2α support retinal angiogenesis by cooperatively maintaining retinal astrocytes in their proangiogenic states. While Tlx+/- and Hif2αf/+/GFAPCre mice are phenotypically normal, Tlx+/-/Hif2αf/+/GFAPCre mice display precocious retinal astrocyte differentiation towards non-angiogenic states, along with significantly reduced retinal angiogenesis. In wild-type mice, TLX and HIF2α coexist in the same protein complex, suggesting a cooperative function under physiological conditions. Furthermore, astrocyte specific disruption of Phd2 (prolyl hydroxylase domain protein 2), a manipulation previously shown to cause HIF2α accumulation, did not rescue retinal angiogenesis in Tlx-/- background, which suggests functional dependence of HIF2α on TLX. Finally, the expression of fibronectin and VEGF-A is significantly reduced in retinal astrocytes of neonatal Tlx+/-/Hif2αf/+/GFAPCre mice. Overall, these data indicate that TLX and HIF2α cooperatively support retinal angiogenesis by maintaining angiogenic potential of retinal astrocytes.

Integrative Analysis of Angiogenesis-Related Long Non-Coding RNA and Identification of a Six-DEARlncRNA Signature Associated with Prognosis and Therapeutic Response in Esophageal Squamous Cell Carcinoma.

Esophageal squamous cell carcinoma (ESCC) is a lethal gastrointestinal malignancy worldwide. We aimed to identify an angiogenesis-related lncRNAs (ARlncRNAs) signature that could predict the prognosis in ESCC. The GSE53624 and GSE53622 datasets were derived from the GEO database. The differently expressed ARlncRNAs (DEARlncRNAs) were retrieved by the weighted gene co-expression network analysis (WGCNA), differential expression analysis, and correlation analysis. Optimal lncRNA biomarkers were screened from the training set and the six-DEARlncRNA signature comprising AP000696.2, LINC01711, RP11-70C1.3, AP000487.5, AC011997.1, and RP11-225N10.1 could separate patients into high- and low-risk groups with markedly different survival. The validation of the reliability of the risk model was performed by the Kaplan-Meier test, ROC curves, and risk curves in the test set and validation set. Predictive independence analysis indicated that risk score is an independent prognostic biomarker for predicting the prognosis of ESCC patients. Subsequently, a ceRNA regulatory network and functional enrichment analysis were performed. The IC50 test revealed that patients in the high-risk group were resistant to Gefitinib and Lapatinib. Finally, the six DEARlncRNAs were detected by qRT-PCR. In conclusion, we demonstrated a novel ARlncRNA signature as an independent prognostic factor to distinguish the risk of ESCC patients and benefit the personalized clinical applications.

Protocol for the development of a clinical practice guideline for discharge planning of stroke patients.

Background: A standardized discharge plan is important to continuous medical care and discharge management of stroke patients. Currently, there is a lack of high-quality, evidence-based discharge planning guidelines for stroke patients. Most existing discharge planning guidelines have been developed for other diseases and stroke-related guidelines focus more on prevention, treatment, and rehabilitation and less on discharge planning. Therefore, they do not provide a systematic and comprehensive answer to the key issues of discharge planning for stroke patients. To improve the level of recovery and quality of life of stroke patients, to better guide clinical caregivers in developing and implementing discharge plans, the Evidence-based Nursing Center of West China Hospital, Sichuan University and the World Health Organization (WHO) Collaborating Centre for Guideline Implementation and Knowledge Translation have jointly initiated the development of the clinical practice guideline for discharge planning of patients with stroke. Methods: The guideline development process is designed to follow the WHO handbook for guideline development and Guidelines 2.0. Evidence grading and guideline recommendations are based on the Grading of Recommendations Assessment, Development, and Evaluation (GRADE). The key steps in developing the guideline include: (I) establishing the guideline working groups; (II) selecting the priority clinical questions; (III) evidence retrieval and evaluation; (IV) grading the quality of evidence; (V) forming recommendations; and (VI) external review. Discussion: This guideline will follow the clinical characteristics and management priorities of stroke and will be developed by a multidisciplinary guideline development team, in strict accordance with the core principles and methods of guideline development. This guideline will provide an evidence-based reference for standardized discharge screening, assessment, discharge procedures, and outpatient follow up, so as to improve the quality of discharge services and standardize the discharge management of stroke patients, and ultimately improve their post-discharge rehabilitation and quality of life. Trial Registration: The guideline was registered at the Practice guidelines REgistration for transPAREncy. The registration No. is IPGRP-2022CN331.

Dependence of Retinal Pigment Epithelium Integrity on the NRF2-Heme Oxygenase-1 Axis.

Purpose: Tight junctions (TJs) form the structural basis of retinal pigment epithelium (RPE) barrier functions. Although oxidative stress contributes to age-related macular degeneration, it is unclear how RPE TJ integrity is controlled by redox balance. In this study, we investigated the protective roles of nuclear factor erythroid 2-related factor 2 (NRF2), a transcription factor, and heme oxygenase-1 (HO1), a heme-degrading enzyme encoded by the NRF2 target gene HMOX1. Methods: ARPE19 cell cultures and mice, including wild-type, Nrf2-/-, and RPE-specific NRF2-deficient mice, were treated with chemicals that impose oxidative stress or impact heme metabolism. In addition, NRF2 and HO1 expression in ARPE19 cells was knocked down by siRNA. TJ integrity was examined by anti-zonula occludens-1 staining of cultured cells or flatmount RPE tissues from mice. RPE barrier functions were evaluated by transepithelium electrical resistance in ARPE19 cells and immunofluorescence staining for albumin or dextran in eye histological sections. Results: TJ structures and RPE barrier functions were compromised due to oxidant exposure and NRF2 deficiency but were rescued by HO1 inducer. Furthermore, treatment with HO1 inhibitor or heme precursor is destructive to TJ structures and RPE barrier properties. Interestingly, both NRF2 and HO1 were upregulated under oxidative stress, probably as an adaptive response to mitigate oxidant-inflicted damages. Conclusions: Our data indicate that the NRF2-HO1 axis protects TJ integrity and RPE barrier functions by driving heme degradation.

Potentialities and Challenges of mRNA Vaccine in Cancer Immunotherapy.

Immunotherapy has become the breakthrough strategies for treatment of cancer in recent years. The application of messenger RNA in cancer immunotherapy is gaining tremendous popularity as mRNA can function as an effective vector for the delivery of therapeutic antibodies on immune targets. The high efficacy, decreased toxicity, rapid manufacturing and safe administration of mRNA vaccines have great advantages over conventional vaccines. The unprecedent success of mRNA vaccines against infection has proved its effectiveness. However, the instability and inefficient delivery of mRNA has cast a shadow on the wide application of this approach. In the past decades, modifications on mRNA structure and delivery methods have been made to solve these questions. This review summarizes recent advancements of mRNA vaccines in cancer immunotherapy and the existing challenges for its clinical application, providing insights on the future optimization of mRNA vaccines for the successful treatment of cancer.

pH-Dependent formation of three porous In(III)-MOFs: framework diversity and selective gas adsorption.

pH-Dependent self-assembly and structural transformation have been observed in a series of porous In(III)-MOFs, H3O[In3(pta)4(OH)2]·10H2O (NXU-1), [In(pta)2]·C3H10N (NXU-2) and [In(pta)2]·C3H10N (NXU-3) (H2pta = 2-(4-pyridyl)-terephthalic acid). The structural diversities of NXU-1-3 reveal that the pH value of the reaction plays a key role in the assembly of In-MOFs. NXU-1 with excellent stability exhibits highly selective CO2 adsorption over CH4 as compared to NXU-2 and NXU-3, owing to the presence of abundant multiple active sites unveiled by theoretical calculations.

An Automatic Method for Epileptic Seizure Detection Based on Deep Metric Learning.

Electroencephalography (EEG) is a commonly used clinical approach for the diagnosis of epilepsy which is a life-threatening neurological disorder. Many algorithms have been proposed for the automatic detection of epileptic seizures using traditional machine learning and deep learning. Although deep learning methods have achieved great success in many fields, their performance in EEG analysis and classification is still limited mainly due to the relatively small sizes of available datasets. In this paper, we propose an automatic method for the detection of epileptic seizures based on deep metric learning which is a novel strategy tackling the few-shot problem by mitigating the demand for massive data. First, two one-dimensional convolutional embedding modules are proposed as a deep feature extractor, for single-channel and multichannel EEG signals respectively. Then, a deep metric learning model is detailed along with a stage-wise training strategy. Experiments are conducted on the publicly-available Bonn University dataset which is a benchmark dataset, and the CHB-MIT dataset which is larger and more realistic. Impressive averaged accuracy of 98.60% and specificity of 100% are achieved on the most difficult classification of interictal (subset D) vs ictal (subset E) of the Bonn dataset. On the CHB-MIT dataset, an averaged accuracy of 86.68% and specificity of 93.71% are reached. With the proposed method, automatic and accurate detection of seizures can be performed in real time, and the heavy burden of neurologists can be effectively reduced.

Oxygen-sensing mechanisms in development and tissue repair.

Under normoxia, hypoxia inducible factor (HIF) α subunits are hydroxylated by PHDs (prolyl hydroxylase domain proteins) and subsequently undergo polyubiquitylation and degradation. Normal embryogenesis occurs under hypoxia, which suppresses PHD activities and allows HIFα to stabilize and regulate development. In this Primer, we explain molecular mechanisms of the oxygen-sensing pathway, summarize HIF-regulated downstream events, discuss loss-of-function phenotypes primarily in mouse development, and highlight clinical relevance to angiogenesis and tissue repair.

A Novel Sleep Staging Network Based on Data Adaptation and Multimodal Fusion.

Sleep staging is one of the important methods to diagnosis and treatment of sleep diseases. However, it is laborious and time-consuming, therefore, computer assisted sleep staging is necessary. Most of the existing sleep staging researches using hand-engineered features rely on prior knowledges of sleep analysis, and usually single channel electroencephalogram (EEG) is used for sleep staging task. Prior knowledge is not always available, and single channel EEG signal cannot fully represent the patient's sleeping physiological states. To tackle the above two problems, we propose an automatic sleep staging network model based on data adaptation and multimodal feature fusion using EEG and electrooculogram (EOG) signals. 3D-CNN is used to extract the time-frequency features of EEG at different time scales, and LSTM is used to learn the frequency evolution of EOG. The nonlinear relationship between the High-layer features of EEG and EOG is fitted by deep probabilistic network. Experiments on SLEEP-EDF and a private dataset show that the proposed model achieves state-of-the-art performance. Moreover, the prediction result is in accordance with that from the expert diagnosis.