Investigation of Chaotic Behaviors of Fractional Order Love Model Without External Environment Effect.

In this paper, we focus on the nonlinear dynamic behavior of fractional order love model because the fractional order can reflect the 'memory dependency' of certain dynamic processes to a certain extent. The novel fractional order love model without external environment effect investigates two aspects: first, the chaotic dynamic of the used system when the system order is 2, and second, the smallest system order of fractional order love model that can generate chaotic behaviors. The simulation results show the fractional order love model can produce different results compared to the integer order model. While the fractional order love model still has chaotic behavior even the sum of the system order is equal to 2. Moreover, the smallest system order of fractional order love model having chaotic behavior is 1.7. The results indicate that two individuals can display love status even if the sum of the system order is less than 2 because the 'memory dependency' effects can greatly affect the emotional changes of human beings. The simulation results based on time series, phase portrait, power spectrum, Poincare map, maximal Lyapunov exponent and bifurcation diagram, and the conclusion is applied to the real life are also discussed.

NUF2 Promotes Breast Cancer Development as a New Tumor Stem Cell Indicator.

Multiple new subtypes of breast cancer (BRCA) are identified in women each year, rendering BRCA the most common and rapidly expanding form of cancer in females globally. NUF2 has been identified as a prognostic factor in various human cancers, regulating cell apoptosis and proliferation. However, its role in BRCA prognosis has not been clarified. This study explored the role of NUF2 in breast cancer development and prognosis using informatic analysis combined with in vivo intracellular studies. Through the online website TIMER, we evaluated the transcription profile of NUF2 across a variety of different cancer types and found that NUF2 mRNA was highly expressed in BRCA patients. Its transcription level was found to be related to the subtype, pathological stage, and prognosis of BRCA. The R program analysis showed a correlation of NUF2 with cell proliferation and tumor stemness in the BRCA patient samples. Subsequently, the association between the NUF2 expression level and immune cell infiltration was analyzed using the XIANTAO and TIMER tools. The results revealed that NUF2 expression was correlated with the responses of multiple immune cells. Furthermore, we observed the effect of NUF2 expression on tumor stemness in BRCA cell lines in vivo. The experimental results illuminated that the overexpression of NUF2 statistically upregulated the proliferation and tumor stemness ability of the BRCA cell lines MCF-7 and Hs-578T. Meanwhile, the knockdown of NUF2 inhibited the abilities of both cell lines, a finding which was verified by analyzing the subcutaneous tumorigenic ability in nude mice. In summary, this study suggests that NUF2 may play a key role in the development and progression of BRCA by affecting tumor stemness. As a stemness indicator, it has the potential to be one of the markers for the diagnosis of BRCA.

[Analysis on quality value transfer of substance benchmarks of Qingwei San].

By preparing 10 batches of substance benchmarks freeze-drying powder( lyophilized powder),the methodology of the characteristic spectrum and the content of index component for substance benchmarks of Qingwei San was established. The characteristic peaks and the similarity range of the characteristic spectrum,the contents and the transfer rate range of isoferulic acid,palmatine and paeonol,and the paste-forming rate range were determined to define key quality attributes of substance benchmarks of Qingwei San. In the10 batches of substance benchmarks of Qingwei San,the similarity of characteristic spectrum was higher than 0. 90. In further comparison of the characteristic peak information,a total of 16 characteristic peaks were identified,including 5 characteristic peaks from Cimicifugae Rhizoma,5 characteristic peaks from Coptidis Rhizoma,2 characteristic peaks from Angelicae Sinensis Radix and 4 characteristic peaks from Moutan Cortex. The content of isoferulic acid was 0. 10%-0. 18%,with the average transfer rate of 49. 82%±4. 02%. The content of palmatine was 0. 17%-0. 31%,with the average transfer rate of 15. 84% ±2. 39%. The content of paeonol was 0. 41%-0. 75%,with the average transfer rate of 23. 41%±3. 23%. The paste-forming rate of the 10 batches of substance benchmarks were controlled at 27%-33%,with the transfer rate between the theoretical paste-forming rate and the actual paste-forming rate was 86. 59%±3. 39%. In this study,the quality value transfer of substance benchmarks of Qingwei San was analyzed by the combination of characteristic spectrum,the content of index component and the paste-forming rate. A scientific and stable evaluation method was preliminarily established,so as to provide the basis for subsequent development and quality control of relevant preparations of Qingwei San.

Design, synthesis and antiproliferative activity evaluation of m-(4-morpholinyl-1,3,5-triazin-2-yl)benzamides in vitro.

In the present study, a series of m-(4-morpholino-1,3,5-triazin-2-yl)benzamides were designed, synthesized and characterized. Their antiproliferative activities against HCT-116 cell and MCF-7 cell at 10µM were evaluated by MTT assay. Compounds T6, T10, T11, T12 and T19 exhibited potent antiproliferative activities. Thus, their IC50 values against HCT-116 cell, MCF-7 cell, Hela cell, U-87 MG cell and A549 cell were measured. The SAR of the target compounds was preliminary discussed. The Western bolt assay suggested that compound T11 can block the PI3K/Akt/mTOR pathway. Hoechst staining assay indicated that compound T11 can cause morphological changes and induce apoptosis of HCT-116 cells. These findings directly identify the m-(4-morpholinyl-1,3,5-triazin-2-yl)benzamide derivatives as novel antiproliferative agents and further verify the value of the benzamide fragment in drug design.

Synthesis and antitumor activity evaluation of PI3K inhibitors containing 3-substituted quinazolin-4(3H)-one moiety.

In present study, a series of N-(2-methoxy-5-(3-substituted quinazolin-4(3H)-one-6-yl)-pyridin-3-yl)phenylsulfonamide were synthesized. Their antiproliferative activities in vitro were evaluated via MTT assay against HCT116 and MCF-7 cancer cell lines. The SAR of title compounds was discussed. The compounds (S)-C5 and (S)-C8 displayed potent inhibitory activity against PI3Ks and mTOR, especially against PI3Kα. In addition, compound (S)-C5 can efficaciously inhibit tumor growth in a mice S-180 model. These findings suggest that our designed compounds can serve as potent PI3K inhibitors and effective anticancer agents.

Modification of N-(6-(2-methoxy-3-(4-fluorophenylsulfonamido)pyridin-5-yl)-[1,2,4]triazolo[1,5-a]pyridin-2-yl)acetamide as PI3Ks inhibitor by replacement of the acetamide group with alkylurea.

N-(6-(2-Methoxy-3-(4-fluorophenylsulfonamido)pyridin-5-yl)-[1,2,4]triazolo[1,5-a]pyridin-2-yl)acetamide exhibits remarkable anticancer effects and toxicity when orally administrated. In present study, alkylurea moiety replaced the acetamide group in the compound and a series of 1-alkyl-3-(6-(2-methoxy-3-sulfonylaminopyridin-5-yl)-[1,2,4]triazolo[1,5-a]pyridin-2-yl)urea derivatives were synthesized. The antiproliferative activities of the synthesized compounds in vitro were evaluated against four human cancer cell lines. Several compounds with potent antiproliferative activities were tested for their acute oral toxicity and their inhibitory activity against PI3Ks and mTOR. The results indicate that the compound attached a alkylurea or 2-(dialkylamino)ethylurea moiety at the 2-position of [1,2,4]triazolo[1,5-a]pyridine can retain the antiproliferative activity and the inhibitory activity against PI3Ks and mTOR. In addition, their acute oral toxicity reduced dramatically. Moreover, the results also indicate that compound 1e can efficaciously inhibit tumor growth in a mice S180 model. These findings suggest that title compounds can serve as potent PI3K inhibitors and effective anticancer agents with low toxicity.

Synthesis and anticancer effects evaluation of 1-alkyl-3-(6-(2-methoxy-3-sulfonylaminopyridin-5-yl)benzo[d]thiazol-2-yl)urea as anticancer agents with low toxicity.

As a PI3K and mTOR dual inhibitor, N-(2-chloro-5-(2-acetylaminobenzo[d]thiazol-6-yl)pyridin-3-yl)-4-fluorophenylsulfonamide displays toxicity when orally administrated. In the present study, alkylurea moiety replaced the acetamide group in the compound and a series of 1-alkyl-3-(6-(2,3-disubstituted pyridin-5-yl)benzo[d]thiazol-2-yl)urea derivatives were synthesized. The antiproliferative activities of the synthesized compounds in vitro were evaluated against HCT116, MCF-7, U87 MG and A549 cell lines. The compounds with potent antiproliferative activity were tested for their acute oral toxicity and inhibitory activity against PI3Ks and mTORC1. The results indicate that the compound attached a 2-(dialkylamino)ethylurea moiety at the 2-positeion of benzothiazole can retain the antiproliferative activity and inhibitory activity against PI3K and mTOR. In addition, their acute oral toxicity reduced dramatically. Moreover, compound 2f can effectively inhibit tumor growth in a mice S180 homograft model. These findings suggest that 1-(2-dialkylaminoethyl)-3-(6-(2-methoxy-3-sulfonylaminopyridin-5-yl)benzo[d]thiazol-2-yl)urea derivatives can serve as potent PI3K inhibitors and anticancer agents with low toxicity.

A FHIR Native Radiology Informatics Platform.

FHIR is a new standard that is rapidly being adopted in healthcare. We describe and implement a Radiology informatics platform (RIS) that is FHIR native and incorporates the ability to execute AI algorithms to aid with the interpretation of scans. Our design utilises the FHIR workflow pattern as an application programming interface with functionality provided by independent micro services thus granting flexibility and expandability.

Knockdown of YTHDF2 initiates ERS-induced apoptosis and cancer stemness suppression by sustaining GLI2 stability in cervical cancer.

Cervical cancer ranks fourth in women in terms of incidence and mortality. The RNA-binding protein YTH N6-methyladenosine RNA-binding protein F2 (YTHDF2) contributes to cancer progression by incompletely understood mechanisms. We show how YTHDF2 controls the fate of cervical cancer cells and whether YTHDF2 could be a valid target for the therapy of cervical cancer. Sphere formation and alkaline phosphatase staining assays were performed to evaluate tumor stemness of cervical cancer cells following YTHDF2 knockdown. Apoptosis was detected by flow cytometry and TUNEL assay. The compounds 4PBA and SP600125 were used to investigate the correlation between JNK, endoplasmic reticulum stress, tumor stemness, and apoptosis. Data from The Cancer Genome Atlas (TCGA) databases and Gene Expression Omnibus (GEO) revealed that GLI family zinc finger 2 (GLI2) might be the target of YTHDF2. The transcription inhibitor actinomycin D and dual-luciferase reporter gene assays were employed to investigate the association between the GLI2 mRNA and YTHDF2. Nude mouse xenografts were generated to assess the effects of YTHDF2 knockdown on cervical cancer growth in vivo. Knockdown of YTHDF2 up-regulated the expression of GLI2, leading to JNK phosphorylation and endoplasmic reticulum stress. These processes inhibited the proliferation of cervical cancer cells and their tumor cell stemness and promotion of apoptosis. In conclusion, the knockdown of YTHDF2 significantly affects the progression of cervical cancer cells, making it a potential target for treating cervical cancer.

Uncertainty estimation using a 3D probabilistic U-Net for segmentation with small radiotherapy clinical trial datasets.

BACKGROUND AND OBJECTIVES: Bio-medical image segmentation models typically attempt to predict one segmentation that resembles a ground-truth structure as closely as possible. However, as medical images are not perfect representations of anatomy, obtaining this ground truth is not possible. A surrogate commonly used is to have multiple expert observers define the same structure for a dataset. When multiple observers define the same structure on the same image there can be significant differences depending on the structure, image quality/modality and the region being defined. It is often desirable to estimate this type of aleatoric uncertainty in a segmentation model to help understand the region in which the true structure is likely to be positioned. Furthermore, obtaining these datasets is resource intensive so training such models using limited data may be required. With a small dataset size, differing patient anatomy is likely not well represented causing epistemic uncertainty which should also be estimated so it can be determined for which cases the model is effective or not. METHODS: We use a 3D probabilistic U-Net to train a model from which several segmentations can be sampled to estimate the range of uncertainty seen between multiple observers. To ensure that regions where observers disagree most are emphasised in model training, we expand the Generalised Evidence Lower Bound (ELBO) with a Constrained Optimisation (GECO) loss function with an additional contour loss term to give attention to this region. Ensemble and Monte-Carlo dropout (MCDO) uncertainty quantification methods are used during inference to estimate model confidence on an unseen case. We apply our methodology to two radiotherapy clinical trial datasets, a gastric cancer trial (TOPGEAR, TROG 08.08) and a post-prostatectomy prostate cancer trial (RAVES, TROG 08.03). Each dataset contains only 10 cases each for model development to segment the clinical target volume (CTV) which was defined by multiple observers on each case. An additional 50 cases are available as a hold-out dataset for each trial which had only one observer define the CTV structure on each case. Up to 50 samples were generated using the probabilistic model for each case in the hold-out dataset. To assess performance, each manually defined structure was matched to the closest matching sampled segmentation based on commonly used metrics. RESULTS: The TOPGEAR CTV model achieved a Dice Similarity Coefficient (DSC) and Surface DSC (sDSC) of 0.7 and 0.43 respectively with the RAVES model achieving 0.75 and 0.71 respectively. Segmentation quality across cases in the hold-out datasets was variable however both the ensemble and MCDO uncertainty estimation approaches were able to accurately estimate model confidence with a p-value < 0.001 for both TOPGEAR and RAVES when comparing the DSC using the Pearson correlation coefficient. CONCLUSIONS: We demonstrated that training auto-segmentation models which can estimate aleatoric and epistemic uncertainty using limited datasets is possible. Having the model estimate prediction confidence is important to understand for which unseen cases a model is likely to be useful.

Bioremediation of the tobacco waste-contaminated soil by Pseudomonas sp. HF-1: nicotine degradation and microbial community analysis.

The highly effective nicotine-degrading bacterium Pseudomonas sp. HF-1 was augmented into the tobacco waste-contaminated soil to degrade nicotine and evaluate the effect of the bioremediation. Comparing with non-adding (NA) systems, the treatments with addition of strain HF-1 (TA) exhibited considerably stronger pollution disposal abilities and higher stability of pH value and moisture content, especially in groups containing a large quantity of tobacco waste. The denaturing gradient gel electrophoresis (DGGE) profiles showed that the Shannon-Wiener index decreased with increasing wastes in the NA treatments, while a gradual increase was found in the TA groups. A comparison of sequences from DGGE bands demonstrated that there were differences in the dominant microbial species between the two treatments, suggesting that strain HF-1 could persist in the soil and enhance the efficiency of tobacco waste disposal. The results of real-time fluorescence quantitative PCR (RT-qPCR) also indicated that strain HF-1 existed in the TA systems and grew with relative high quantities. In conclusion, the nicotine-degrading strain HF-1 played a leading role in the bioremediation of the tobacco waste-contaminated soil and influenced the dynamics and structure of the microbial community.

HBx promotes hepatocellular carcinoma progression by repressing the transcription level of miR-187-5p.

HBV-associated hepatitis B virus x protein (HBx) plays multiple roles in the development of hepatocellular carcinoma. In our prior study, we discovered that miR-187-5p expression was inhibited by HBx. To investigate the underlying molecular mechanism of HBx-mediated miR-187-5p downregulation in hepatocellular carcinoma cells, effects of HBx and miR-187-5p on hepatoma carcinoma cell were observed, as well as their interactions. Through in vitro and in vivo experiments, we demonstrated that overexpression of miR-187-5p inhibited proliferation, migration, and invasion. Simultaneously, we observed a dysregulation in the expression of miR-187-5p in liver cancer cell lines, which may be attributed to transcriptional inhibition through the E2F1/FoxP3 axis. Additionally, we noted that HBx protein is capable of enhancing the expression of E2F1, a transcription factor that promotes the expression of FoxP3. In conclusion, our results suggest that the inhibitory effect of HBx on miR-187-5p is mediated through the E2F1/FoxP3 axis. As shown in this work, HBx promotes hepatoma carcinoma cell proliferation, migration, and invasion through the E2F1/FoxP3/miR-187 axis. It provides a theoretical basis for finding therapeutic targets that will help clinic treatment for HCC.

Automatic classification of distal radius fracture using a two-stage ensemble deep learning framework.

Distal radius fractures (DRFs) are one of the most common types of wrist fracture and can be subdivided into intra- and extra-articular fractures. Compared with extra-articular DRFs which spare the joint surface, intra-articular DRFs extend to the articular surface and can be more difficult to treat. Identification of articular involvement can provide valuable information about the characteristics of fracture patterns. In this study, a two-stage ensemble deep learning framework was proposed to differentiate intra- and extra-articular DRFs automatically on posteroanterior (PA) view wrist X-rays. The framework firstly detects the distal radius region of interest (ROI) using an ensemble model of YOLOv5 networks, which imitates the clinicians' search pattern of zooming in on relevant regions to assess abnormalities. Secondly, an ensemble model of EfficientNet-B3 networks classifies the fractures in the detected ROIs into intra- and extra-articular. The framework achieved an area under the receiver operating characteristic curve of 0.82, an accuracy of 0.81, a true positive rate of 0.83 and a false positive rate of 0.27 (specificity of 0.73) for differentiating intra- from extra-articular DRFs. This study has demonstrated the potential in automatic DRF characterization using deep learning on clinically acquired wrist radiographs and can serve as a baseline for further research in incorporating multi-view information for fracture classification.

A Food Waste-Derived Organic Liquid Fertiliser for Sustainable Hydroponic Cultivation of Lettuce, Cucumber and Cherry Tomato.

We previously reported a sustainable food waste management approach to produce an acceptable organic liquid fertiliser for recycling food waste called "FoodLift." This study follows our previous work to evaluate the macronutrients and cation concentrations in harvested structural parts of lettuce, cucumber, and cherry tomatoes produced using food waste-derived liquid fertiliser (FoodLift) and compare them against commercial liquid fertiliser (CLF) under hydroponic conditions. N and P concentrations in the structural parts of lettuce and the fruit and plant structural parts of cucumber appear to be similar between FoodLift and CLF (p > 0.05), with significantly different N concentrations in the various parts of cherry tomato plants (p < 0.05). For lettuce, N and P content varied from 50 to 260 g/kg and 11 to 88 g/kg, respectively. For cucumber and cherry tomato plants, N and P concentrations ranged from 1 to 36 g/kg and 4 to 33 g/kg, respectively. FoodLift was not effective as a nutrient source for growing cherry tomatoes. Moreover, the cation (K, Ca, and Mg) concentrations appear to significantly differ between FoodLift and CLF grown plants (p < 0.05). For example, for cucumber, Ca content varied from 2 to 18 g/kg for FoodLift grown plants while Ca in CLF-grown cucumber plants ranged from 2 to 28 g/kg. Overall, as suggested in our previous work, FoodLift has the potential to replace CLF in hydroponic systems for lettuce and cucumber. This will lead to sustainable food production, recycling of food waste to produce liquid fertiliser, and will promote a circular economy in nutrient management.

Paraformaldehyde as C1 Synthon: Electrochemical Three-Component Synthesis of Tetrahydroimidazo[1,5-a]quinoxalin-4(5H)-ones in Aqueous Ethanol.

A green and practical method for the electrochemical synthesis of tetrahydroimidazo[1,5-a]quinoxalin-4(5H)-ones through the three-component reaction of quinoxalin-2(1H)-ones, N-arylglycines and paraformaldehyde was reported. In this strategy, EtOH played dual roles (eco-friendly solvent and waste-free pre-catalyst) and the in situ generated ethoxide promoted triple sequential deprotonations.

Clinical target volume delineation quality assurance for MRI-guided prostate radiotherapy using deep learning with uncertainty estimation.

BACKGROUND AND PURPOSE: Previous studies on automatic delineation quality assurance (QA) have mostly focused on CT-based planning. As MRI-guided radiotherapy is increasingly utilized in prostate cancer treatment, there is a need for more research on MRI-specific automatic QA. This work proposes a clinical target volume (CTV) delineation QA framework based on deep learning (DL) for MRI-guided prostate radiotherapy. MATERIALS AND METHODS: The proposed workflow utilized a 3D dropblock ResUnet++ (DB-ResUnet++) to generate multiple segmentation predictions via Monte Carlo dropout which were used to compute an average delineation and area of uncertainty. A logistic regression (LR) classifier was employed to classify the manual delineation as pass or discrepancy based on the spatial association between the manual delineation and the network's outputs. This approach was evaluated on a multicentre MRI-only prostate radiotherapy dataset and compared with our previously published QA framework based on AN-AG Unet. RESULTS: The proposed framework achieved an area under the receiver operating curve (AUROC) of 0.92, a true positive rate (TPR) of 0.92 and a false positive rate of 0.09 with an average processing time per delineation of 1.3 min. Compared with our previous work using AN-AG Unet, this method generated fewer false positive detections at the same TPR with a much faster processing speed. CONCLUSION: To the best of our knowledge, this is the first study to propose an automatic delineation QA tool using DL with uncertainty estimation for MRI-guided prostate radiotherapy, which can potentially be used for reviewing prostate CTV delineation in multicentre clinical trials.

Ferrocene/air double-mediated FeTiO3-photocatalyzed semi-heterogeneous annulation of quinoxalin-2(1H)-ones in EtOH/H2O.

With both ferrocene and air as the redox catalysts, for the first time, the low-cost natural ilmenite (FeTiO3) was successfully used for photocatalytic bond formations. Under the assistance of a traceless H-bond, and HCHO as the methylene reagent, a variety of imidazo[1,5-a]quinoxalinones were semi-heterogeneously photosynthesized in high yields with good functional group compatibility.

Detection of ETV6 gene rearrangements in adult acute lymphoblastic leukemia.

ETV6 is an important hematopoietic regulatory factor and ETV6 gene rearrangement is involved in a wide variety of hematological malignancies. In this study, we sought to investigate the incidence of ETV6-associated fusion genes in B- and T-lineage acute lymphoblastic leukemia (ALL) by multiplex-nested reverse transcription-polymerase chain reaction (RT-PCR) in 176 adult ALL patients. Total RNA was extracted from bone marrow samples of ALL patients including 136 B- and 40 T-lineage ALL, and ETV6 fusion genes were detected by multiplex-nested RT-PCR. Changes of ETV6 fusion gene mRNA transcript levels were examined by real-time RT-PCR. We detected a total of 15 ETV6 gene rearrangements with a positive rate of 8.5%, involving seven ETV6-associated fusion genes in 13 B-ALL (13/136, 9.6%) and 2 T-ALL patients (2/40, 5.0%). ETV6-RUNX1 were observed in six cases (3.4%), ETV6-JAK2 in three cases (1.7%), ETV6-ABL1 in two cases (1.1%), and ETV6-ABL2, ETV6-NCOA2, ETV6-SYK, and PAX5-ETV6 each in one case (0.6%). ETV6-JAK2 was found in both B-ALL and T-ALL patients. Furthermore, real-time quantitative RT-PCR assays showed that the ETV6-RUNX1 mRNA transcript levels decreased during conventional chemotherapy or hematopoietic stem cell transplantation. This study shows that multiplex-nested RT-PCR is an effective and accurate tool to identify ETV6 rearrangements in adult ALL, which provides some clues into the diagnosis and prognosis of ALL but also molecular markers for the detection of minimal residual disease in adult ALL.

The effect of tetrahydrofuran on the enzymatic activity and microbial community in activated sludge from a sequencing batch reactor.

Tetrahydrofuran (THF) is a toxic and carcinogenic compound that is commonly released from pharmaceutical, chemical and related industry wastewater. Currently, the effects of THF contamination on wastewater are unknown and a better understanding of THF toxicity toward biological processes in wastewater treatment is critical. In this study, we firstly investigated the toxic effects of THF on enzymatic activity and the microbial diversity in activated sludge from a sequencing batch reactor during long-term exposure to 10 mM THF. The activity of five enzymes (catalase, dehydrogenase, urease, phosphatase and protease) was remarkably decreased in the presence of 10 mM THF during a period of 85 days. Of these five affected enzymes, dehydrogenase activity was close to detection level limits and was nearly completely inhibited. Analysis of the microbial community demonstrated that THF, at a concentration of 10 mM, altered the distribution of microbes within the community and significantly decreased microbial diversity during long-term contamination, according to denaturing gradient gel electrophoresis (DGGE) analysis. The fraction of Actinobacteria increased in the community, while the fraction of Proteobacteria significantly decreased after THF exposure.

Lambda-vector modeling temporal and channel interactions for text-independent speaker verification.

Most of the current excellent models in speaker verification are ResNet-based deep models and attention-based models. These models have a general weakness, which is the large number of parameters and high hardware requirements. On the other hand, many deep structures only generate embedding features from the features extracted by the last frame-level layer, which causes shallow features and channel-related features to be ignored. To solve these problems, this paper proposed a shallow speaker verification model based on Lambda-vector, its main structure is composed of three Lambda-SE modules. The module extracts long-distance dependencies between frame-level features and channel-related interaction information to enhance representation of features. Meanwhile, so that adequately mine the information in deep and shallow features, the model introduces multi-layer feature aggregation to fuse the features of different frame-level layers together. It can increase the detailed information in the deep features and improve the model's ability to represent complex information. The experimental results on the public datasets Voxceleb1 and Voxceleb2 show that the model has more stable training speed, fewer model parameters, and better identification performances than baseline models.