Malignancy-associated ischemic stroke: Implications for diagnostic and therapeutic workup.

BACKGROUND: Patients with malignancies have an increased risk of suffering ischemic stroke via several mechanisms such as coagulation dysfunction and other malignancy-related effects as well as iatrogenic causes. Moreover, stroke can be the first sign of an occult malignancy, termed as malignancy-associated ischemic stroke (MAS). Therefore, timely diagnostic assessment and targeted management of this complex clinical situation are critical. FINDINGS: Patients with both stroke and malignancy have atypical ages, risk factors, and often exhibit malignancy-related symptoms and multiple lesions on neuroimaging. New biomarkers such as eicosapentaenoic acid and blood mRNA profiles may help in distinguishing MAS from other strokes. In terms of treatment, malignancy should not be considered a contraindication, given comparable rates of recanalization and complications between stroke patients with or without malignancies. CONCLUSION: In this review, we summarize the latest developments in diagnosing and managing MAS, especially stroke with occult malignancies, and provide new recommendations from recently emerged clinical evidence for diagnostic and therapeutic workup strategies.

Development and Validation of a Deep-Learning Network for Detecting Congenital Heart Disease from Multi-View Multi-Modal Transthoracic Echocardiograms.

Early detection and treatment of congenital heart disease (CHD) can significantly improve the prognosis of children. However, inexperienced sonographers often face difficulties in recognizing CHD through transthoracic echocardiogram (TTE) images. In this study, 2-dimensional (2D) and Doppler TTEs of children collected from 2 clinical groups from Beijing Children's Hospital between 2018 and 2022 were analyzed, including views of apical 4 chamber, subxiphoid long-axis view of 2 atria, parasternal long-axis view of the left ventricle, parasternal short-axis view of aorta, and suprasternal long-axis view. A deep learning (DL) framework was developed to identify cardiac views, integrate information from various views and modalities, visualize the high-risk region, and predict the probability of the subject being normal or having an atrial septal defect (ASD) or a ventricular septaldefect (VSD). A total of 1,932 children (1,255 healthy controls, 292 ASDs, and 385 VSDs) were collected from 2 clinical groups. For view classification, the DL model reached a mean [SD] accuracy of 0.989 [0.001]. For CHD screening, the model using both 2D and Doppler TTEs with 5 views achieved a mean [SD] area under the receiver operating characteristic curve (AUC) of 0.996 [0.000] and an accuracy of 0.994 [0.002] for within-center evaluation while reaching a mean [SD] AUC of 0.990 [0.003] and an accuracy of 0.993 [0.001] for cross-center test set. For the classification of healthy, ASD, and VSD, the model reached the mean [SD] accuracy of 0.991 [0.002] and 0.986 [0.001] for within- and cross-center evaluation, respectively. The DL models aggregating TTEs with more modalities and scanning views attained superior performance to approximate that of experienced sonographers. The incorporation of multiple views and modalities of TTEs in the model enables accurate identification of children with CHD in a noninvasive manner, suggesting the potential to enhance CHD detection performance and simplify the screening process.

Burden of intracerebral haemorrhage in Europe: forecasting incidence and mortality between 2019 and 2050.

Background: Anticipating the burden of intracerebral haemorrhage is crucial for proactive management and building resilience against future health challenges. Prior forecasts are based on population demography and to a lesser extent epidemiological trends. This study aims to utilise selected modifiable risk factors and socio-demographic indicators to forecast the incidence and mortality of intracerebral haemorrhage in Europe between 2019 and 2050. Methods: Three intracerebral haemorrhage risk factors identified in the Global Burden of Diseases, Injuries, and Risk Factors study (GBD 2019)-high systolic blood pressure, high fasting plasma glucose, and high body mass index-were utilised to predict the risk-attributable fractions between 2019 and 2050. Disease burden not attributable to these risk factors was then forecasted using time series models (autoregressive integrated moving average [ARIMA]), incorporating the Socio-demographic Index (SDI) as an external predictor. The optimal parameters of ARIMA models were selected for each age-sex-country group based on the Akaike Information Criterion (AIC). Different health scenarios were constructed by extending the past 85th and 15th percentiles of annualised rates of change in risk factors and SDI across all location-years, stratified by age and sex groups. A decomposition analysis was performed to assess the relative contributions of population size, age composition, and intracerebral haemorrhage risk on the projected changes. Findings: Compared with observed figures in 2019, our analysis predicts an increase in the burden of intracerebral haemorrhage in Europe in 2050, with a marginal rise of 0.6% (95% uncertainty interval [UI], -7.4% to 9.6%) in incident cases and an 8.9% (-2.8% to 23.6%) increase in mortality, reaching 141.2 (120.6-166.5) thousand and 144.2 (122.9-172.2) thousand respectively. These projections may fluctuate depending on trajectories of the risk factors and SDI; worsened trends could result in increases of 16.7% (8.7%-25.3%) in incidence and 31.2% (17.7%-48%) in mortality, while better trajectories may lead to a 10% (16.4%-2.3%) decrease in intracerebral haemorrhage cases with stabilised mortality. Individuals aged ≥80 years are expected to contribute significantly to the burden, comprising 62.7% of the cases in 2050, up from 40% in 2019, and 72.5% of deaths, up from 50.5%. Country-wide variations were noted in the projected changes, with decreases in the standardised rates across all nations but varying crude rates. The largest relative reductions in counts for both incidence and mortality are expected in Latvia, Bulgaria, and Hungary-ranging from -38.2% to -32.4% and -37.3% to -30.2% respectively. In contrast, the greatest increases for both measures were forecasted in Ireland (45.7% and 74.4%), Luxembourg (45% and 70.7%), and Cyprus (44.5% and 74.2%). The modelled increase in the burden of intracerebral haemorrhage could largely be attributed to population ageing. Interpretation: This study provides a comprehensive forecast of intracerebral haemorrhage in Europe until 2050, presenting different trajectories. The potential increase in the number of people experiencing and dying from intracerebral haemorrhage could have profound implications for both caregiving responsibilities and associated costs. However, forecasts were divergent between different scenarios and among EU countries, signalling the pivotal role of public health initiatives in steering the trajectories. Funding: The European Union's Horizon 2020 Research and Innovation Programme under grant agreement No. 754517. The National Institute for Health and Care Research (NIHR) under its Programme Grants for Applied Research (NIHR202339).

Deep learning-based lesion detection and severity grading of small-bowel Crohn's disease ulcers on double-balloon endoscopy images.

BACKGROUND AND AIMS: Double-balloon endoscopy (DBE) is widely used in diagnosing small-bowel Crohn's disease (CD). However, CD misdiagnosis frequently occurs if inexperienced endoscopists cannot accurately detect the lesions. The CD evaluation may also be inaccurate owing to the subjectivity of endoscopists. This study aimed to use artificial intelligence (AI) to accurately detect and objectively assess small-bowel CD for more refined disease management. METHODS: We collected 28,155 small-bowel DBE images from 628 patients from January 2018 to December 2022. Four expert gastroenterologists labeled the images, and at least 2 endoscopists made the final decision with agreement. A state-of-the-art deep learning model, EfficientNet-b5, was trained to detect CD lesions and evaluate CD ulcers. The detection included lesions of ulcer, noninflammatory stenosis, and inflammatory stenosis. Ulcer grading included ulcerated surface, ulcer size, and ulcer depth. A comparison of AI model performance with endoscopists was performed. RESULTS: The EfficientNet-b5 achieved high accuracies of 96.3% (95% confidence interval [CI], 95.7%-96.7%), 95.7% (95% CI, 95.1%-96.2%), and 96.7% (95% CI, 96.2%-97.2%) for the detection of ulcers, noninflammatory stenosis, and inflammatory stenosis, respectively. In ulcer grading, the EfficientNet-b5 exhibited average accuracies of 87.3% (95% CI, 84.6%-89.6%) for grading the ulcerated surface, 87.8% (95% CI, 85.0%-90.2%) for grading the size of ulcers, and 85.2% (95% CI, 83.2%-87.0%) for ulcer depth assessment. CONCLUSIONS: The EfficientNet-b5 achieved high accuracy in detecting CD lesions and grading CD ulcers. The AI model can provide expert-level accuracy and objective evaluation of small-bowel CD to optimize the clinical treatment plans.

Source-free domain adaptive segmentation with class-balanced complementary self-training.

Unsupervised domain adaptation (UDA) plays a crucial role in transferring knowledge gained from a labeled source domain to effectively apply it in an unlabeled and diverse target domain. While UDA commonly involves training on data from both domains, accessing labeled data from the source domain is frequently constrained, citing concerns related to patient data privacy or intellectual property. The source-free UDA (SFUDA) can be promising to sidestep this difficulty. However, without the source domain supervision, the SFUDA methods can easily fall into the dilemma of "winner takes all", in which the majority category can dominate the deep segmentor, and the minority categories are largely ignored. In addition, the over-confident pseudo-label noise in self-training-based UDA is a long-lasting problem. To sidestep these difficulties, we propose a novel class-balanced complementary self-training (CBCOST) framework for SFUDA segmentation. Specifically, we jointly optimize the pseudo-label-based self-training with two mutually reinforced components. The first class-wise balanced pseudo-label training (CBT) explicitly exploits the fine-grained class-wise confidence to select the class-wise balanced pseudo-labeled pixels with the adaptive within-class thresholds. Second, to alleviate the pseudo-labeled noise, we propose a complementary self-training (COST) to exclude the classes that do not belong to, with a heuristic complementary label selection scheme. We evaluated our CBCOST framework on both 2D and 3D cross-modality cardiac anatomical segmentation tasks and brain tumor segmentation tasks. Our experimental results showed that our CBCOST performs better than existing SFUDA methods and yields similar performance, compared with UDA methods with the source data.

Construction and evaluation of a predictive model for compassion fatigue among emergency department nurses: A cross-sectional study.

BACKGROUND: Compassion fatigue is a syndrome resulting from long-term work-related traumatic event stress exposure of medical staff. The emergency department is considered to be a high-risk, high-intensity and high-stress work environment, with a high prevalence of trauma and violence. Nurses in the emergency department are more prone to compassion fatigue than nurses in other departments. Compassion fatigue not only affects the physical and mental health, and job satisfaction of emergency department nurses, but also causes serious consequences for patients, such as poor patient outcome, medical errors, and increased patient mortality during hospitalization. OBJECTIVES: Our study aims to develop and evaluate a predictive model for compassion fatigue among emergency department nurses. DESIGN: A cross-sectional study. DATA SOURCES: The emergency department nurses (N = 1014) were recruited from 21 tertiary hospitals (from Chengdu, Chongqing, Guiyang, Guangzhou and Shanghai) in central, southwestern, southern, and eastern China from July 25, 2022 to October 30, 2022. METHODS: Univariate and multiple logistic regression analyses were used to determine the potential predictive factors associated with compassion fatigue in emergency department nurses. A nomogram was built based on the predictive factors and internally evaluated using a bootstrap resampling method (1000 bootstrap resamples). The performance of the predictive model was evaluated by measuring the Hosmer-Lemeshow goodness of fit test and calibration curve. RESULTS: The prevalence of compassion fatigue among emergency department nurses was 75.9 %. The multiple logistic regression analysis revealed that the independent predictive factors for compassion fatigue among emergency department nurses were working position, job satisfaction, diet habit, sleep hours per day, occupational stress, physical harassment and the level of workplace violence, all of which were identified to create the nomogram. The Hosmer-Lemeshow goodness of fit test indicated that the predictive model was well calibrated (χ2 = 11.520, P = 0.174). The bootstrap-corrected concordance index of nomogram was 0.821 (95 % CI: 0.791-0.851). The calibration curve of the nomogram showed good consistency between the predicted and actual probabilities. CONCLUSIONS: A predictive model of compassion fatigue among emergency department nurses has been developed, based on the general demographic, work-related and lifestyle characteristics, occupational stress, and workplace violence, with satisfactory predictive ability. This model can identify emergency department nurses who are at high risk of compassion fatigue. Our study provides an empirical basis for early detection, early diagnosis and early intervention of emergency department nurses at high risk of compassion fatigue.

[Effect of Pulsating Vacuum Cleaning Sterilizer on the Cleaning Quality of Dental Handpieces].

Objective: To explore the effect of pulsating vacuum cleaning sterilizer on the cleaning quality of dental handpieces. Methods: A total of 390 newly-purchased high-speed dental handpieces were covered in the study. By the random number table method, the handpieces were divided into three groups that were cleaned by different methods-Group A ( n=130), pulsating vacuum cleaning sterilizer, group B ( n=130), automatic thermal cleaning and disinfection machine, and group C ( n=130), manual cleaning. The cleaning quality, internal cavity drying and the length of cleaning time of the three groups were compared. Results: The cleaning quality of group A (100%) was better than those of groups B (89.2%) and C (83.8%) and the length of cleaning time of group A (1.13 h) was shorter than those of group B (1.80 h) and C (2.60 h), all showing significant differences ( P<0.05). In addition, the cleaning quality of group B was better than that of group C and the length of cleaning time of group B was shorter than that of group C, all showing significant differences ( P<0.05). However, there was no significant difference in the internal cavity drying effects between the three groups ( P>0.05). Conclusions: The pulsating vacuum cleaning sterilizer can effectively improve the cleaning quality of dental handpieces and shorten the cleaning time. Hence, it should be extensively applied and promoted in clinic service.

Chemokine receptor CXCR7 activates Aurora Kinase A and promotes neuroendocrine prostate cancer growth.

CXCR7 is an atypical chemokine receptor that recruits ß-arrestin (ARRB2) and internalizes into clathrin-coated intracellular vesicles where the complex acts as a scaffold for cytoplasmic kinase assembly and signal transduction. Here, we report that CXCR7 was elevated in the majority of prostate cancer (PCa) cases with neuroendocrine features (NEPC). CXCR7 markedly induced mitotic spindle and cell cycle gene expression. Mechanistically, we identified Aurora Kinase A (AURKA), a key regulator of mitosis, as a novel target that was bound and activated by the CXCR7-ARRB2 complex. CXCR7 interacted with proteins associated with microtubules and golgi, and, as such, the CXCR7-ARRB2-containing vesicles trafficked along the microtubules to the pericentrosomal golgi apparatus, where the complex interacted with AURKA. Accordingly, CXCR7 promoted PCa cell proliferation and tumor growth, which was mitigated by AURKA inhibition. In summary, our study reveals a critical role of CXCR7-ARRB2 in interacting and activating AURKA, which can be targeted by AURKA inhibitors to benefit a subset of patients with NEPC.

Neutrophil-derived cathelicidin promotes cerebral angiogenesis after ischemic stroke.

Neutrophils play critical roles in the evolving of brain injuries following ischemic stroke. However, how they impact the brain repair in the late phase after stroke remain uncertain. Using a prospective clinical stroke patient cohort, we found significantly increased cathelicidin antimicrobial peptide (CAMP) in the peripheral blood of stroke patients compared to that of healthy controls. While in the mouse stroke model, CAMP was present in the peripheral blood, brain ischemic core and significantly increased at day 1, 3, 7, 14 after middle cerebral artery occlusion (MCAO). CAMP-/- mice exhibited significantly increased infarct volume, exacerbated neurological outcome, reduced cerebral endothelial cell proliferation and vascular density at 7 and 14 days after MCAO. Using bEND3 cells subjected to oxygen-glucose deprivation (OGD), we found significantly increased angiogenesis-related gene expression with the treatment of recombinant CAMP peptide (rCAMP) after reoxygenation. Intracerebroventricular injection (ICV) of AZD-5069, the antagonist of CAMP receptor CXCR2, or knockdown of CXCR2 by shCXCR2 recombinant adeno-associated virus (rAAV) impeded angiogenesis and neurological recovery after MCAO. Administration of rCAMP promoted endothelial proliferation and angiogenesis and attenuated neurological deficits 14 days after MCAO. In conclusion, neutrophil derived CAMP represents an important mediator that could promote post-stroke angiogenesis and neurological recovery in the late phase after stroke.

Ferroptosis: the vulnerability within a cancer monster.

Treatment-resistant cancer, such as neuroendocrine prostate cancer (NEPC), is a lethal disease with limited therapeutic options. RB1 is a tumor suppressor gene that is lost in a majority of NEPC tumors. In this issue of the JCI, Wang and colleagues examined how RB1 loss may sensitize cancer cells to ferroptosis inducers through elevation of ACSL4, a key enzyme that promotes lipid peroxidation and triggers ferroptosis. We discuss a high potential of RB1-deficient cells to undergo ferroptosis due to the elevation of ACSL4. This is normally kept in check by abundant expression of GPX4, an antioxidant enzyme, in cancer cells. This balance, however, is tilted by GPX4 inhibitors, leading to massive ferroptosis. We highlight possible therapeutic strategies that exploit this inherent vulnerability for targeting RB1-deficient, treatment-resistant cancer.

Psychological capital and organizational citizenship behavior among nurses during the COVID-19 epidemic: mediation of organizational commitment.

BACKGROUND: Nurses' organizational citizenship behavior, a spontaneous "altruistic work behavior", may be affected by psychological capital and organizational commitment, but its mechanism is not clear. The aim of this study was to investigate the characteristics and distribution of psychological capital, organizational commitment and organizational citizenship behavior among nurses during the COVID-19 epidemic, and explore the mediating role of organizational commitment in psychological capital and organizational citizenship behavior. METHODS: A cross-sectional survey was conducted among 746 nurses from 6 designated hospitals for COVID-19 treatment in China. Descriptive statistics, Pearson correlation analysis, and structural equation model were used in this study. RESULTS: Nurses' psychological capital, organizational commitment and organizational citizenship behavior scores were 103.12 ± 15.57, 46.53 ± 7.14 and 101.47 ± 12.14, respectively. Additionally, organizational commitment partially mediates between psychological capital and organizational citizenship behavior. CONCLUSIONS: During the COVID-19 pandemic, nurses' psychological capital, organizational commitment, and organizational citizenship behavior were found to be at an upper-middle level, influenced by various social-demographic factors. Furthermore, the results illustrated that psychological capital can affect organizational citizenship behavior through the mediating role of organizational commitment. Therefore, the findings emphasize the importance of nursing administration to monitor and prioritize the mental health and organizational behavior of nurses during the ongoing COVID-19 crisis. It is crucial to focus on developing and nurturing nurses' psychological capital, strengthening their organizational commitment, and ultimately promoting their organizational citizenship behavior.

Perspective Review of Myeloid Immune Cell Responses and Poststroke Immunosuppression.

Ischemic stroke profoundly influences the peripheral immune system, which responds quickly to brain ischemia and participates in the evolution of poststroke neuroinflammation, while a period of systemic immunosuppression ensues. Poststroke immunosuppression brings harmful consequences, including increased infection rates and escalated death. As the most abundant cell population in the fast-responding innate immune system, myeloid cells including neutrophils and monocytes play an indispensable role in systemic immunosuppression after stroke. The change in myeloid response after stroke can be regulated by circulating DAMPs (damage-associated molecular patterns) and neuromodulatory mechanisms, which contain sympathetic nervous system, hypothalamic-pituitary-adrenal, and parasympathetic nervous system. In this review, we summarize the emerging roles and newly identified mechanisms underlying myeloid cell response in poststroke immunosuppression. Deeper understanding of the above points may pave the way for future development of novel therapeutic strategies to treat poststroke immunosuppression.

The levels and related factors of mental workload among nurses: A systematic review and meta-analysis.

AIM: The aim was to determine the overall levels and related factors of mental workload assessed using the NASA-TLX tool among nurses. BACKGROUND: Mental workload is a key element that affects nursing performance. However, there exists no review regarding mental workload assessed using the NASA-TLX tool, focusing on nurses. DESIGN: A systematic review and meta-analysis. DATA SOURCES: PubMed, MEDLINE, Web of Science, EMBASE, PsycINFO, Scopus, CINAHL, CNKI, CBM, Weipu and WanFang databases were searched from 1 January 1998 to 30 February 2022. REVIEW METHODS: Following the PRISMA statement recommendations, review methods resulted in 31 quantitative studies retained for inclusion which were evaluated with the evaluation criteria for observational studies as recommended by the Agency for Healthcare Research and Quality. The data were pooled and a random-effects meta-analysis conducted. RESULTS: Findings showed the pooled mental workload score was 65.24, and the pooled prevalence of high mental workload was 54%. Subgroup analysis indicated nurses in developing countries and emergency departments experienced higher mental workloads, and the mental workloads of front-line nurses increased significantly during the COVID-19 pandemic. CONCLUSION: These findings highlight that nurses experience high mental workloads as assessed using the NASA-TLX tool and there is an urgent need to explore interventions to decrease their mental workloads.

Dualistic roles and mechanistic insights of macrophage migration inhibitory factor in brain injury and neurodegenerative diseases.

Macrophage migration inhibitory factor (MIF) is involved in various immune-mediated pathologies and regulates both innate and adaptive immune reactions, thus being related to several acute and chronic inflammatory diseases such as rheumatoid arthritis, septic shock, and atherosclerosis. Its role in acute and chronic brain pathologies, such as stroke and neurodegenerative diseases, has attracted increasing attention in recent years. In response to stimuli like hypoxia, inflammation or infection, different cell types can rapidly release MIF, including immune cells, endothelial cells, and neuron cells. Notably, clinical data from past decades also suggested a possible link between serum MIF levels and the severity of stroke and the evolving of neurodegenerative diseases. In this review, we summarize the major and recent findings focusing on the mechanisms of MIF modulating functions in brain injury and neurodegenerative diseases, which may provide important therapeutic targets meriting further investigation.

Assessment of Transcatheter or Surgical Closure of Atrial Septal Defect using Interpretable Deep Keypoint Stadiometry.

Automated echocardiogram interpretation with artificial intelligence (AI) has the potential to facilitate the serial diagnosis of heart defects by primary clinician. However, the fully automated and interpretable analysis pipeline for suggesting a treatment plan is largely underexplored. The present study targets to build an automatic and interpretable assistant for the transthoracic echocardiogram- (TTE-) based assessment of atrial septal defect (ASD) with deep learning (DL). We developed a novel deep keypoint stadiometry (DKS) model, which learns to precisely localize the keypoints, i.e., the endpoints of defects and followed by the absolute distance measurement with the scale. The closure plan and the size of the ASD occluder for transcatheter closure are derived based on the explicit clinical decision rules. A total of 3,474 2D and Doppler TTE from 579 patients were retrospectively collected from two clinical groups. The accuracy of closure classification using DKS (0.9425 ± 0.0052) outperforms the "black-box" model (0.7646 ± 0.0068; p < 0.0001) for within-center evaluation. The results in cross-center cases or using the quadratic weighted kappa as an evaluation metric are consistent. The fine-grained keypoint label provides more explicit supervision for network training. While DKS can be fully automated, clinicians can intervene and edit at different steps of the process as well. Our deep learning keypoint localization can provide an automatic and transparent way for assessing size-sensitive congenital heart defects, which has huge potential value for application in primary medical institutions in China. Also, more size-sensitive treatment planning tasks may be explored in the future.

The relationship between organizational commitment and work engagement among clinical nurses in China: A cross-sectional study.

AIM: This study aims to investigate the levels of organizational commitment and work engagement among clinical nurses in tertiary hospitals and explore the relationship between them. BACKGROUND: The nursing literature supports the idea that organizational commitment plays an important role in positively influencing job performance. However, the relationship between organizational commitment and work engagement among clinical nurses remains unclear. METHODS: This was a cross-sectional study. A convenience sample of clinical nurses (n = 621) was selected from five tertiary hospitals in Sichuan Province of China. Survey instruments included a general information questionnaire about organizational commitment and work engagement. Univariate analysis, correlation analyses and linear regression analysis were used to examine the association between organizational commitment and work engagement. RESULTS: The mean scores for organizational commitment and work engagement were 3.85 ± 0.59 and 4.58 ± 1.46, respectively. A moderate degree of positive correlation was found between them. Gender, monthly income and retention commitment were significantly associated with work engagement, and they accounted for 39.0% of the total variance. CONCLUSIONS: Clinical nurses had a moderate level of organizational commitment and a high level of work engagement. Organizational commitment positively influenced work engagement; that is, organizational commitment is a significant determinant of nurses' work engagement. IMPLICATIONS FOR NURSING MANAGEMENT: Hospital organizations should focus on assessing and strengthening nurses' organizational commitment to promote increased work engagement and, ultimately, improved quality of care. This may include, but is not limited to, increasing rest time for nurses, implementing a performance appraisal system and focusing on nurses' psychological state.

Prevalence and factors of posttraumatic growth among Hubei residents during the COVID-19 pandemic: A cross-sectional study.

The adverse impact of the COVID-19 pandemic on mental health has been widely concerned, but the research on positive psychological factors is insufficient, this study aims to investigate the prevalence and factors of posttraumatic growth (PTG) among residents in the worst-hit areas of China (Hubei Province). We were conducted in three disaster areas with different severity in Hubei Province three months after the outbreak, and the data were from 575 respondents. Instruments included the simplified Chinese version of the Posttraumatic Growth Inventory (C-PTGI) and the public health emergency psychological questionnaire. Nonparametric tests, Spearman correlation analyses, and multiple linear regression equations were used to analyze the data. The results showed that three months after the outbreak of COVID-19, the PTG of Hubei residents was at a low level, and their sense of fear was the most prominent, with a positive detection rate of 82.09%. According to the results of this study, high PTG levels were associated with high levels of fear, married and healthcare providers, while low levels of PTG were associated with serious chronic diseases, males, good economic status, and poor prevention and control measures. Government departments should strengthen social support for residents, carry out necessary stress management training to help them correctly deal with negative emotions and promote their personal growth.

Multimodal fusion diagnosis of depression and anxiety based on CNN-LSTM model.

BACKGROUND: In recent years, more and more people suffer from depression and anxiety. These symptoms are hard to be spotted and can be very dangerous. Currently, the Self-Reported Anxiety Scale (SAS) and Self-Reported Depression Scale (SDS) are commonly used for initial screening for depression and anxiety disorders. However, the information contained in these two scales is limited, while the symptoms of subjects are various and complex, which results in the inconsistency between the questionnaire evaluation results and the clinician's diagnosis results. To fully mine the scale data, we propose a method to extract the features from the facial expression and movements, which are generated from the video recorded simultaneously when subjects fill in the scale. Then we collect the facial expression, movements and scale information to establish a multimodal framework for improving the accuracy and robustness of the diagnosis of depression and anxiety. METHODS: We collect the scale results of the subjects and the videos when filling in the scales. Given the two scales, SAS and SDS, we construct a model with two branches, where each branch processes the multimodal data of SAS and SDS, respectively. In the branch, we first build a convolutional neural network (CNN) to extracts the facial expression features in each frame of images. Secondly, we establish a long short-term memory (LSTM) network to further embedding the facial expression feature and build the connections between various frames, so that the movement feature in the video can be generated. Thirdly, we transform the scale scores into one-hot format, and feed them into the corresponding branch of the network to further mining the information of the multimodal data. Finally, we fuse the embeddings of these two branches to generate inference results of depression and anxiety. RESULTS AND CONCLUSIONS: Based on the score results of SAS and SDS, our multimodal model further mines the video information, and can reach the accuracy of 0.946 in diagnosing depression and anxiety. This study demonstrates the feasibility of using our CNN-LSTM-based multimodal model for initial screening and diagnosis of depression and anxiety disorders with high diagnostic performance.