In audiovisual emotion recognition, representational learning is a research direction receiving considerable attention, and the key lies in constructing effective affective representations with both consistency and variability. However, there are still many challenges to accurately realize affective representations. For this reason, in this paper we proposed a cross-modal audiovisual recognition model based on a multi-head cross-attention mechanism. The model achieved fused feature and modality alignment through a multi-head cross-attention architecture, and adopted a segmented training strategy to cope with the modality missing problem. In addition, a unimodal auxiliary loss task was designed and shared parameters were used in order to preserve the independent information of each modality. Ultimately, the model achieved macro and micro F1 scores of 84.5% and 88.2%, respectively, on the crowdsourced annotated multimodal emotion dataset of actor performances (CREMA-D). The model in this paper can effectively capture intra- and inter-modal feature representations of audio and video modalities, and successfully solves the unity problem of the unimodal and multimodal emotion recognition frameworks, which provides a brand-new solution to the audiovisual emotion recognition.
Emotions ; Humans ; Attention ; Algorithms

Automatic classification of medical questions is of great significance in improving the quality and efficiency of online medical services, and belongs to the task of intent recognition. Joint entity recognition and intent recognition perform better than single task models. Currently, most publicly available medical text intent recognition datasets lack entity annotation, and manual annotation of these entities requires a lot of time and manpower. To solve this problem, this paper proposes a medical text classification model, bidirectional encoder representation based on transformer-recurrent convolutional neural network-entity-label-semantics (BRELS), which integrates medical entity label semantics. This model firstly utilizes an adaptive fusion mechanism to absorb prior knowledge of medical entity labels, achieving local feature enhancement. Then in global feature extraction, a lightweight recurrent convolutional neural network (LRCNN) is used to suppress parameter growth while preserving the original semantics of the text. The ablation and comparison experiments are conducted on three public medical text intent recognition datasets to validate the performance of the model. The results show that F1 score reaches 87.34%, 81.71%, and 77.74% on each dataset, respectively. The results show that the BRELS model can effectively identify and understand medical terminology, thereby effectively identifying users' intentions, which can improve the quality and efficiency of online medical services.
Semantics ; Neural Networks, Computer ; Humans ; Natural Language Processing

Deactivation of the mitochondrial pyruvate dehydrogenase complex (PDC) is important for the metabolic switching of cancer cell from oxidative phosphorylation to aerobic glycolysis. Studies examining PDC activity regulation have mainly focused on the phosphorylation of pyruvate dehydrogenase (E1), leaving other post-translational modifications largely unexplored. Here, we demonstrate that the acetylation of Lys 488 of pyruvate dehydrogenase complex component X (PDHX) commonly occurs in hepatocellular carcinoma, disrupting PDC assembly and contributing to lactate-driven epigenetic control of gene expression. PDHX, an E3-binding protein in the PDC, is acetylated by the p300 at Lys 488, impeding the interaction between PDHX and dihydrolipoyl transacetylase (E2), thereby disrupting PDC assembly to inhibit its activation. PDC disruption results in the conversion of most glucose to lactate, contributing to the aerobic glycolysis and H3K56 lactylation-mediated gene expression, facilitating tumor progression. These findings highlight a previously unrecognized role of PDHX acetylation in regulating PDC assembly and activity, linking PDHX Lys 488 acetylation and histone lactylation during hepatocellular carcinoma progression and providing a potential biomarker and therapeutic target for further development.
Humans ; Acetylation ; Carcinoma, Hepatocellular/genetics* ; Liver Neoplasms/genetics* ; Pyruvate Dehydrogenase Complex/genetics* ; Gene Expression Regulation, Neoplastic ; Animals ; Mice ; Cell Line, Tumor ; Protein Processing, Post-Translational ; Histones/metabolism* ; Disease Progression

A complete plant body consists of elements on different scales, including microscopic molecules, mesoscopic multicellular structures, and macroscopic tissues and organs, which are interconnected to form complex biological networks. The growth and development of plants involve the regulation of elements on different scales and their biological networks, which requires the coordinated operation of multiple molecules, cells, tissues, and organs. It is difficult to reveal the essence of multi-level life activities by a single method or technology. In recent years, the development of various novel imaging technologies has provided new approaches for revealing the complex life activities in plants. Using multi-modal imaging technologies to study the cross-scale network connections of plants from the microscopic, mesoscopic, and macroscopic levels is crucial for understanding the complex internal connections behind biological functions. This paper first summarizes multi-modal cross-scale imaging technologies, three-dimensional reconstruction, and image processing methods, outlines the basic framework of cross-scale network connection properties, and then summarizes the applications of multi-modal imaging technologies in elucidating plant multi-scale networks. Finally, this review systematically integrates the combined analysis of cross-scale 3D spatial structural data and single-cell omics, laying a theoretical foundation for the innovation of novel plant imaging technologies. Furthermore, it provides a new research paradigm for in-depth exploration of the interaction mechanisms among cross-scale elements and the principles of biological network connectivity in plant life activities.
Plants/metabolism* ; Imaging, Three-Dimensional/methods* ; Image Processing, Computer-Assisted/methods* ; Multimodal Imaging/methods* ; Plant Physiological Phenomena

ObjectiveTo investigate the syndrome evolution patterns， characteristics of the used herbal medicinals， and efficacy variations across different stages of rheumatoid arthritis （RA） progression. MethodsBased on the China Rheumatoid Arthritis Registry of Patients with Chinese Medicine （CERTAIN）， 2,506 RA patients were retrospectively enrolled and categorized into ＜6 months group （166 cases）， 6 months to ＜5 years group （1063 cases）， 5 to ＜20 years group （1067 cases）， and ≥20 years group （210 cases）. Syndromes were differentiated before and after traditional Chinese medicine （TCM） treatment， including damp-heat obstruction， wind-damp obstruction， cold-damp obstruction， blood stasis obstructed in the collaterals， phlegm-stasis obstruction， liver-kidney insufficiency， qi and blood deficiency， and qi-yin deficiency. The syndrome evolution rate was calculated for high-frequency syndromes before and after treatment. Analysis was conducted on top 20 frequently used Chinese herbs at the first diagnosis. Clinical efficacy of the 28-joint disease activity score based on erythrocyte sedimentation rate （DAS28-ESR） and 28-joint disease activity score based on C-reactive protein （DAS28-CRP） before and after treatment were assessed. A multivariate logistic regression analysis was performed to identify factors affecting the efficacy of TCM treatment. ResultsPatients with course of disease shorter than 6 months predominantly presented with cold-dampness obstruction syndrome （49/166， 29.5%）， wind-dampness obstruction syndrome （46/166， 27.7%）， and dampness-heat obstruction syndrome （43/166， 25.9%）. For patients with course of disease logner than 6 months and shorter than 5 years and those within 5 to 20 years， the dominant syndrome was dampness-heat obstruction syndrome （324/1063， 30.5% and 318/1067， 29.8%， respectively）. In patients with disease duration ≥20 years， liver-kidney insufficiency syndrome and dampness-heat obstruction syndrome both predominated， each accounting for 25.24% （53/210）. The syndromes with more than 100 cases before treatment and a syndrome evolution rate greater than 10% after treatment were dampness-heat obstruction （201/738， 27.2%）， liver-kidney insufficiency （119/367， 32.4%）， and phlegm-stasis obstruction syndromes （73/172， 42.4%）. These were classified as high-frequency syndromes. After treatment， damp-heat obstruction syndrome and liver-kidney insufficiency syndrome primarily evolved into wind-damp obstruction syndrome， while phlegm-stasis obstruction syndrome evolved into damp-heat obstruction and cold-damp obstruction syndrome. The top two commonly used Chinese herbs across all groups were Gancao （Radix et Rhizoma Glycyrrhizae） and Baishao （Radix Paeoniae Alba）. In the ＜6 months group and the 6 months to ＜5 years group， high-frequency herbs also included Fangfeng （Radix Saposhnikoviae）， Duhuo （Radix Angelicae Pubescentis）， Chuanxiong （Rhizoma Chuanxiong）， and Qianghuo （Radix et Rhizoma Notopterygii）. In the 5 to ＜20 years group and the ≥20 years group， the usage of Huangqi （Radix Astragali）， Fuling （Poria）， Niuxi （Radix Achyranthis Bidentatae）， and Danggui （Radix Angelicae Sinensis） increased， while the proportion of Fangfeng and Duhuo decreased. After treatment， the DAS28-ESR and DAS28-CRP scores in all groups significantly decreased （P＜0.05）. There were statistically significant differences in clinical efficacy based on DAS28-ESR and DAS28-CRP across all groups （P＜0.01）. Multivariate logistic regression revealed significantly reduced treatment efficacy in the 6 months-5 years group （OR=0.4）， 5~20 years group （OR=0.5）， and ≥20 years group （OR=0.4） compared to the ＜6 months group. ConclusionRA syndromes follow a progression pattern from excess to deficiency， with corresponding transition in herbal usage from pathogen-eliminating to healthy qi-reinforcing approaches. TCM intervention can significantly reduce disease activity of RA， with superior efficacy in patients with disease duration shorter than 6 months.

ObjectiveTo evaluate the clinical efficacy and safety of Tripterygium wilfordii polyglycoside tablets （TWP） in the treatment of rheumatoid arthritis （RA） in the real world. MethodDiagnosis and treatment data of patients with RA in Chinese medicine rheumatology registration research information platform information database （CERTAIN） from January 1，2019 to January， 2024 were collected. According to the inclusion and exclusion criteria， data were screened. The included data were divided into an exposure group and a control group according to the use of TWP or not. Propensity score matching （PSM） was used in both groups to keep the baseline balanced and comparable. The disease activity score （DAS28-ESR） of 28 joints based on the erythrocyte sedimentation rate （ESR）before and after treatment was compared between the two groups， as well as health assessment questionnaire （HAQ），visual analogue scale （VAS），tender joint count （TJC），swollen joint count （SJC）， patient's global assessment （PGA），evaluator's global assessment （EGA），laboratory indexes， clinical curative effect， and adverse reactions. ResultA total of 3 978 patients were included，including 929 in the exposure group and 3 049 in the control group. Before PSM，there were significant differences in demographic information，DAS28-ESR score，PGA，EGA，HAQ，VAS scores，SJC， and TJC between the two groups （P<0.05）. After successful PSM matching，922 patients in the exposure group and 922 patients in the control group were included. There was no significant difference in demographic information and DAS28-ESR between the two groups before treatment，and the differences in other indexes between the two groups decreased to varying degrees. After treatment，the DAS28-ESR，PGA，EGA，HAQ，SJC，TJC，VAS scores， ESR，and IgG immune index of the two groups were significantly lower （P<0.01）. Compared with those in the control group after treatment，the DAS28-ESR，PGA，EGA，HAQ，VAS scores， and ESR in the exposure group after treatment decreased more significantly （P<0.05，P<0.01）. There was no significant difference in TJC in the exposure group after treatment. However， TJC in the exposure group was significantly higher than that in the control group before treatment （P<0.05）. In terms of TJC reduction，the exposure group performed better than the control group. There was no significant difference in SJC and IgG between the exposure group and the control group after treatment. After treatment，the clinical symptoms of poor appetite，insomnia and many dreams，upset，fatigue，and fear of wind and cold in the two groups were improved. Except that the proportion of women in the exposure group was higher than that in the control group （P<0.01），there was no significant difference in the incidence of other adverse reactions between the two groups after treatment. ConclusionTWP to treat RA can effectively reduce DAS28-ESR，PGA，EGA，HAQ，TJC，and VAS scores and improve the general symptoms. Except for the women at childbearing age with fertility requirements that TWP is not applicable，it shows good security.

Objective:To evaluate the feasibility of using two-dimensional shear wave elastography (2D-SWE) based liver and spleen elastic hardness (L/S-SWE) in patients with liver cirrhosis, and to determine the exclusion and diagnostic thresholds for early identification of liver cirrhosis.Methods:A total of 574 patients with chronic hepatitis B (hepatitis B for short) were included in this study. The clinical characteristics, L-SWE and S-SWE of the patients were collected, and the differences between cirrhosis group ( n=311) and non cirrhosis group ( n=263) were analyzed. The success rate and stability of liver and spleen elastic surgery were evaluated in two groups. The receiver operating characteristic (ROC) curve was used to analyze the efficacy of L-SWE, S-SWE, aspartate aminotransferase to platelet ratio index (APRI) alone and in combination in diagnosing liver cirrhosis. By analyzing the ROC curve, the double threshold for excluding and diagnosing liver cirrhosis was determined. Results:There was a statistically significant difference in platelet count and APRI between the cirrhosis group and the non cirrhosis group (all P<0.05). In the feasibility assessment of 2D-SWE technology, the success rate and stability of liver and spleen elastic operation were relatively high (success rate: 97.2% vs 81.3%; stability: 0.92 vs 0.84), and the success rate and stability of L-SWE operation were slightly better than S-SWE. The success rate of S-SWE operation in the cirrhosis group was higher than that in the non cirrhosis group ( P<0.05). The correlation analysis results showed that L-SWE, S-SWE, APRI were positively correlated with liver tissue pathological grading ( r=0.677, 0.528, 0.149, all P<0.05). The areas under the ROC curve for identifying liver cirrhosis using L-SWE, S-SWE, and APRI were 0.959, 0.896, and 0.706, respectively. When L-SWE and S-SWE were combined, the area under the ROC curve was 0.987, the sensitivity was 92.6%, and the specificity was 96.0%. The Delong test showed that the combined diagnosis of L-SWE and S-SWE had the same diagnostic efficacy as using L-SWE alone for liver cirrhosis ( P>0.05). Further analysis of the ROC curve showed that the likelihood of liver cirrhosis was low when L-SWE was less than 9.4 kPa, and high when L-SWE was greater than 12.0 kPa. Patients between 9.4 and 12.0 kPa can undergo further S-SWE testing; If the S-SWE was between 17.5 and 29.3 kPa, it was classified as 2D-SWE, which was difficult to determine whether there was liver cirrhosis, and further liver puncture and other examinations were needed. Conclusions:2D-SWE technology has high operational feasibility in the diagnosis of liver cirrhosis, and combined with S-SWE, it helps to improve the diagnostic efficiency of early non-invasive identification of liver cirrhosis, enabling more patients to avoid unnecessary liver puncture examinations.

BACKGROUND:Trauma-induced coagulopathy(TIC)due to serious injuries significantly leads to increased mortality and morbidity among elderly patients.However,the risk factors of TIC are not well elucidated.This study aimed to explore the risk factors of TIC in elderly patients who have major trauma. METHODS:In this retrospective study,the risk factors for TIC in elderly trauma patients at a single trauma center were investigated between January 2015 and September 2020.The demographic information including gender,age,trauma parts,injury severity,use of blood products,use of vasopressors,need of emergency surgery,duration of mechanical ventilation,length of stay in the intensive care unit(ICU)and hospital,and clinical outcomes were extracted from electric medical records.Multivariate logistic regression analysis was performed to differentiate risk factors,and the performance of the model was evaluated using receiver operating characteristics(ROC)curves. RESULTS:Among the 371 elderly trauma patients,248(66.8%)were male,with the age of 72.5±6.8 years,median injury severity score(ISS)of 24(IQR:17-29),and Glasgow coma score(GCS)of 14(IQR:7-15).Of these patients,129(34.8%)were diagnosed with TIC,whereas 242(65.2%)were diagnosed with non-TIC.The severity scores such as ISS(25[20-34]vs.21[16-29],P<0.001)and shock index(SI),(0.90±0.66 vs.0.58±0.18,P<0.001)was significantly higher in the TIC group than in the non-TIC group.Serum calcium levels(1.97±0.19 mmol/L vs.2.15±0.16 mmol/L,P<0.001),fibrinogen levels(1.7±0.8 g/L vs.2.8±0.9 g/L,P<0.001),and base excess(BE,-4.9±4.6 mmol/L vs.-1.2±3.1 mmol/L,P<0.001)were significantly lower in the TIC group than in the non-TIC group.Multivariate logistic regression analysis revealed that ISS>16(OR:3.404,95%CI:1.471-7.880;P=0.004),SI>1(OR:5.641,95%CI:1.700-18.719;P=0.005),low BE(OR:0.868,95%CI:0.760-0.991;P=0.037),hypocalcemia(OR:0.060,95%CI:0.009-0.392;P=0.003),and hypofibrinogenemia(OR:0.266,95%CI:0.168-0.419;P<0.001)were independent risk factors for TIC in elderly trauma patients.The AUC of the prediction model included all these risk factors was 0.887(95%CI:0.851-0.923)with a sensitivity and specificity of 83.6%and 82.6%,respectively. CONCLUSION:Higher ISS(more than 16),higher SI(more than 1),acidosis,hypocalcemia,and hypofibrinogenemia emerged as independent risk factors for TIC in elderly trauma patients.

BACKGROUND:Trauma-induced coagulopathy(TIC)due to serious injuries significantly leads to increased mortality and morbidity among elderly patients.However,the risk factors of TIC are not well elucidated.This study aimed to explore the risk factors of TIC in elderly patients who have major trauma. METHODS:In this retrospective study,the risk factors for TIC in elderly trauma patients at a single trauma center were investigated between January 2015 and September 2020.The demographic information including gender,age,trauma parts,injury severity,use of blood products,use of vasopressors,need of emergency surgery,duration of mechanical ventilation,length of stay in the intensive care unit(ICU)and hospital,and clinical outcomes were extracted from electric medical records.Multivariate logistic regression analysis was performed to differentiate risk factors,and the performance of the model was evaluated using receiver operating characteristics(ROC)curves. RESULTS:Among the 371 elderly trauma patients,248(66.8%)were male,with the age of 72.5±6.8 years,median injury severity score(ISS)of 24(IQR:17-29),and Glasgow coma score(GCS)of 14(IQR:7-15).Of these patients,129(34.8%)were diagnosed with TIC,whereas 242(65.2%)were diagnosed with non-TIC.The severity scores such as ISS(25[20-34]vs.21[16-29],P<0.001)and shock index(SI),(0.90±0.66 vs.0.58±0.18,P<0.001)was significantly higher in the TIC group than in the non-TIC group.Serum calcium levels(1.97±0.19 mmol/L vs.2.15±0.16 mmol/L,P<0.001),fibrinogen levels(1.7±0.8 g/L vs.2.8±0.9 g/L,P<0.001),and base excess(BE,-4.9±4.6 mmol/L vs.-1.2±3.1 mmol/L,P<0.001)were significantly lower in the TIC group than in the non-TIC group.Multivariate logistic regression analysis revealed that ISS>16(OR:3.404,95%CI:1.471-7.880;P=0.004),SI>1(OR:5.641,95%CI:1.700-18.719;P=0.005),low BE(OR:0.868,95%CI:0.760-0.991;P=0.037),hypocalcemia(OR:0.060,95%CI:0.009-0.392;P=0.003),and hypofibrinogenemia(OR:0.266,95%CI:0.168-0.419;P<0.001)were independent risk factors for TIC in elderly trauma patients.The AUC of the prediction model included all these risk factors was 0.887(95%CI:0.851-0.923)with a sensitivity and specificity of 83.6%and 82.6%,respectively. CONCLUSION:Higher ISS(more than 16),higher SI(more than 1),acidosis,hypocalcemia,and hypofibrinogenemia emerged as independent risk factors for TIC in elderly trauma patients.

BACKGROUND:Trauma-induced coagulopathy(TIC)due to serious injuries significantly leads to increased mortality and morbidity among elderly patients.However,the risk factors of TIC are not well elucidated.This study aimed to explore the risk factors of TIC in elderly patients who have major trauma. METHODS:In this retrospective study,the risk factors for TIC in elderly trauma patients at a single trauma center were investigated between January 2015 and September 2020.The demographic information including gender,age,trauma parts,injury severity,use of blood products,use of vasopressors,need of emergency surgery,duration of mechanical ventilation,length of stay in the intensive care unit(ICU)and hospital,and clinical outcomes were extracted from electric medical records.Multivariate logistic regression analysis was performed to differentiate risk factors,and the performance of the model was evaluated using receiver operating characteristics(ROC)curves. RESULTS:Among the 371 elderly trauma patients,248(66.8%)were male,with the age of 72.5±6.8 years,median injury severity score(ISS)of 24(IQR:17-29),and Glasgow coma score(GCS)of 14(IQR:7-15).Of these patients,129(34.8%)were diagnosed with TIC,whereas 242(65.2%)were diagnosed with non-TIC.The severity scores such as ISS(25[20-34]vs.21[16-29],P<0.001)and shock index(SI),(0.90±0.66 vs.0.58±0.18,P<0.001)was significantly higher in the TIC group than in the non-TIC group.Serum calcium levels(1.97±0.19 mmol/L vs.2.15±0.16 mmol/L,P<0.001),fibrinogen levels(1.7±0.8 g/L vs.2.8±0.9 g/L,P<0.001),and base excess(BE,-4.9±4.6 mmol/L vs.-1.2±3.1 mmol/L,P<0.001)were significantly lower in the TIC group than in the non-TIC group.Multivariate logistic regression analysis revealed that ISS>16(OR:3.404,95%CI:1.471-7.880;P=0.004),SI>1(OR:5.641,95%CI:1.700-18.719;P=0.005),low BE(OR:0.868,95%CI:0.760-0.991;P=0.037),hypocalcemia(OR:0.060,95%CI:0.009-0.392;P=0.003),and hypofibrinogenemia(OR:0.266,95%CI:0.168-0.419;P<0.001)were independent risk factors for TIC in elderly trauma patients.The AUC of the prediction model included all these risk factors was 0.887(95%CI:0.851-0.923)with a sensitivity and specificity of 83.6%and 82.6%,respectively. CONCLUSION:Higher ISS(more than 16),higher SI(more than 1),acidosis,hypocalcemia,and hypofibrinogenemia emerged as independent risk factors for TIC in elderly trauma patients.