Association between preoperative anxiety states and postoperative complications in patients with esophageal cancer and COPD: a retrospective cohort study

BACKGROUND

Esophageal cancer brings emotional changes, especially anxiety to patients. Co-existing anxiety makes the surgery difficult and may cause complications. This study aims to evaluate effects of anxiety in postoperative complications of esophageal cancer patients with chronic obstructive pulmonary disease (COPD).

METHODS

Patients with esophageal cancer and co-existing COPD underwent tumor excision. Anxiety was measured using Hospital Anxiety and Depression Scale (HAD) before surgery. Clavien-Dindo criteria were used to grade surgical complications. A multiple regression model was used to analyze the relationship between anxiety and postoperative complications. The chi-square test was used to compare the differences in various types of complications between the anxiety group and the non-anxiety group. A multinomial logistic regression model was used to analyze the influencing factors of mild and severe complications.

RESULTS

This study included a total of 270 eligible patients, of which 20.7% had anxiety symptoms and 56.6% experienced postoperative complications. After evaluation by univariate analysis and multivariate logistic regression models, the risk of developing complications in anxious patients was 4.1 times than non-anxious patients. Anxious patients were more likely to develop pneumonia, pyloric obstruction, and arrhythmia. The presence of anxiety, surgical method, higher body mass index (BMI), and lower preoperative oxygen pressure may increase the incidence of minor complications. The use of surgical methods, higher COPD assessment test (CAT) scores, and higher BMI may increase the incidence of major complications, while anxiety does not affect the occurrence of major complications (P = 0.054).

CONCLUSION

Preoperative anxiety is associated with postoperative complications in esophageal cancer patients with co-existing COPD. Anxiety may increase the incidence of postoperative complications, especially minor complications in patient with COPD and esophageal cancer.

Neural network establishes co-occurrence links between transformation products of the contaminant and the soil microbiome

It remains challenging to establish reliable links between transformation products (TPs) of contaminants and corresponding microbes. This challenge arises due to the sophisticated experimental regime required for TP discovery and the compositional nature of 16S rRNA gene amplicon sequencing and mass spectrometry datasets, which can potentially confound statistical inference. In this study, we present a new strategy by combining the use of 2H-labeled Stable Isotope-Assisted Metabolomics (2H-SIAM) with a neural network-based algorithm (i.e., MMvec) to explore links between TPs of pyrene and the soil microbiome. The links established by this novel strategy were further validated using different approaches. Briefly, a metagenomic study provided indirect evidence for the established links, while the identification of pyrene degraders from soils, and a DNA-based stable isotope probing (DNA-SIP) study offered direct evidence. The comparison among different approaches, including Pearson's and Spearman's correlations, further confirmed the superior performance of our strategy. In conclusion, we summarize the unique features of the combined use of 2H-SIAM and MMvec. This study not only addresses the challenges in linking TPs to microbes but also introduces an innovative and effective approach for such investigations. Environmental Implication: Taxonomically diverse bacteria performing successive metabolic steps of the contaminant were firstly depicted in the environmental matrix.

Solving the non-submodular network collapse problems via Decision Transformer

Given a graph G, the network collapse problem (NCP) selects a vertex subset S of minimum cardinality from G such that the difference in the values of a given measure function f(G)-f(G∖S) is greater than a predefined collapse threshold. Many graph analytic applications can be formulated as NCPs with different measure functions, which often pose a significant challenge due to their NP-hard nature. As a result, traditional greedy algorithms, which select the vertex with the highest reward at each step, may not effectively find the optimal solution. In addition, existing learning-based algorithms do not have the ability to model the sequence of actions taken during the decision-making process, making it difficult to capture the combinatorial effect of selected vertices on the final solution. This limits the performance of learning-based approaches in non-submodular NCPs. To address these limitations, we propose a unified framework called DT-NC, which adapts the Decision Transformer to the Network Collapse problems. DT-NC takes into account the historical actions taken during the decision-making process and effectively captures the combinatorial effect of selected vertices. The ability of DT-NC to model the dependency among selected vertices allows it to address the difficulties caused by the non-submodular property of measure functions in some NCPs effectively. Through extensive experiments on various NCPs and graphs of different sizes, we demonstrate that DT-NC outperforms the state-of-the-art methods and exhibits excellent transferability and generalizability.

The longitudinal behavioral effects of acute exposure to galactic cosmic radiation in female C57BL/6J mice: implications for deep space missions, female crews, and potential antioxidant countermeasures

Galactic cosmic radiation (GCR) is an unavoidable risk to astronauts that may affect mission success. Male rodents exposed to 33-beam-GCR (33-GCR) show short-term cognitive deficits but reports on female rodents and long-term assessment is lacking. Here we asked: What are the longitudinal behavioral effects of 33-GCR on female mice? Also, can an antioxidant/anti-inflammatory compound mitigate the impact of 33-GCR? Mature (6-month-old) C57BL/6J female mice received the antioxidant CDDO-EA (400 µg/g of food) or a control diet (vehicle, Veh) for 5 days and either Sham-irradiation (IRR) or whole-body 33-GCR (0.75Gy) on the 4th day. Three-months post-IRR, mice underwent two touchscreen-platform tests: 1) location discrimination reversal (which tests behavior pattern separation and cognitive flexibility, two abilities reliant on the dentate gyrus) and 2) stimulus-response learning/extinction. Mice then underwent arena-based behavior tests (e.g. open field, 3-chamber social interaction). At the experiment end (14.25-month post-IRR), neurogenesis was assessed (doublecortin-immunoreactive [DCX+] dentate gyrus neurons). Female mice exposed to Veh/Sham vs. Veh/33-GCR had similar pattern separation (% correct to 1st reversal). There were two effects of diet: CDDO-EA/Sham and CDDO-EA/33-GCR mice had better pattern separation vs. their respective control groups (Veh/Sham, Veh/33-GCR), and CDDO-EA/33-GCR mice had better cognitive flexibility (reversal number) vs. Veh/33-GCR mice. Notably, one radiation effect/CDDO-EA countereffect also emerged: Veh/33-GCR mice had worse stimulus-response learning (days to completion) vs. all other groups, including CDDO-EA/33-GCR mice. In general, all mice show normal anxiety-like behavior, exploration, and habituation to novel environments. There was also a change in neurogenesis: Veh/33-GCR mice had fewer DCX+ dentate gyrus immature neurons vs. Veh/Sham mice. Our study implies space radiation is a risk to a female crew's longitudinal mission-relevant cognitive processes and CDDO-EA is a potential dietary countermeasure for space-radiation CNS risks.

Metabolomic profiling of serum and tongue coating of pregnant women with intrahepatic cholestasis in pregnancy

Intrahepatic cholestasis of pregnancy (ICP) is associated with an increased risk of cesarean section and adverse fetal outcomes. Currently, ICP diagnosis depends largely on serum levels of bile acids and lacks sensitivity and specificity for accurate diagnosis. Tongue diagnosis is an important diagnostic tool in traditional Chinese medicine (TCM) and is used in our clinic as complementary treatment and personalized medicine for ICP. However, the molecular basis of the manifestation of greasy white tongue coatings in ICP remains unknown. In this study, we performed untargeted metabolomic profiling of the serum, tongue coating, and saliva of 66 pregnant women, including 22 with ICP. The metabolomic profiles of the serum and tongue coatings showed marked differences between the two clinical groups. Forty-six differentially abundant metabolites were identified, and their relative concentrations correlated with total bile acid levels. These differential metabolites included bile acids, lipids, microbiota- and diet-related metabolites, and exposomes. Conventional biochemical markers, including serum aminotransferases and bilirubin, were not significantly increased in the ICP group, whereas the total cholesterol and triglyceride levels were significantly increased as early as the first trimester. Our data provide insights into the pathophysiology of ICP and implicate the gut-liver axis and environmental exposure. Tongue coating has the potential to be a non-invasive diagnostic approach. Further studies are required to validate the clinical utility of these findings.

scPROTEIN: a versatile deep graph contrastive learning framework for single-cell proteomics embedding

Single-cell proteomics sequencing technology sheds light on protein-protein interactions, posttranslational modifications and proteoform dynamics in the cell. However, the uncertainty estimation for peptide quantification, data missingness, batch effects and high noise hinder the analysis of single-cell proteomic data. It is important to solve this set of tangled problems together, but the existing methods tailored for single-cell transcriptomes cannot fully address this task. Here we propose a versatile framework designed for single-cell proteomics data analysis called scPROTEIN, which consists of peptide uncertainty estimation based on a multitask heteroscedastic regression model and cell embedding generation based on graph contrastive learning. scPROTEIN can estimate the uncertainty of peptide quantification, denoise protein data, remove batch effects and encode single-cell proteomic-specific embeddings in a unified framework. We demonstrate that scPROTEIN is efficient for cell clustering, batch correction, cell type annotation, clinical analysis and spatially resolved proteomic data exploration.

Reduced contrast sensitivity function is correlated with changes to cone photoreceptors in simple high myopia

Purpose

To investigate the contrast sensitivity function (CSF) changes in simple high myopia (SHM) and evaluate the correlations between these changes with the early changes in the retinal microstructure.

Methods

This prospective study comprised 81 subjects, 20 with emmetropia (EM), 26 with low myopia and moderate myopia (LM/MM), and 35 with SHM. The area under the log CSF curve (AULCSF) and the cut-off spatial frequency (Cut-off SF) were employed as measures of CSF. Adaptive optics (AO) was employed to quantify the cone density, spacing, and regularity. The thickness and blood flow of the retinal sublayers were determined from vertical and horizontal optical coherence tomography angiography (OCTA) A-scans. Swept-source optical coherence tomography (SS-OCT) was employed to analyze the choroidal thickness (CT) and choroidal vascularity using a custom algorithm. Differences in the retinal and choroidal parameters, cone distribution, AULCSF, and Cut-off SF were compared among the three groups. Multivariate linear mixed models were used to elucidate the associations between photoreceptor morphological alterations, retinal and choroidal parameters, and AULCSF.

Results

The AULCSF and Cut-off SF were significantly lower in the SHM group compared to the EM and LM groups (p < 0.05). The SHM group had less cone density, larger cone spacing, and lower cone regularity than the EM and LM/MM groups (p < 0.05). Moreover, the thickness of the inner segment of photoreceptors (IS), retinal pigment epithelium (RPE) layer and choroid were reduced, and the outer segment of photoreceptors (OS) was thicker in the SHM group compared to the EM and LM/MM groups (all p < 0.05). A longer axial length (AL) was correlated with decreased AULCSF, cone density, and cone spacing (r = -0.800 to 0.752, all p < 0.050). Additionally, decreased CSF was correlated with lower cone density (r = 0.338, p = 0.035).

Conclusion

Decreased contrast sensitivity was observed in patients with SHM and cone density was significantly correlated with reduced AUCSF.

Survival benefits of adjuvant chemotherapy for patients with residual pathologic disease after neoadjuvant chemotherapy and surgery for locally advanced esophageal squamous cell carcinoma

BACKGROUND

Although preoperative neoadjuvant chemotherapy (NACT) or chemoradiation is the current standard of care for esophageal cancer in China, the impact of subsequent adjuvant therapy on patient prognosis remains unknown. This study aims to analyze the effect of adjuvant chemotherapy (ACT) on the survival rates of patients who have achieved a non-pathological complete response (non-pCR) after NACT and subsequent surgery.

METHODS

We reviewed the data of 2193 patients with locally advanced thoracic esophageal squamous cell carcinoma (ESCC) who underwent radical surgery between January 2006 and January 2016. Of these patients, 46 received NACT and ACT, while 109 received NACT only. Propensity score matching was used to compare 86 patients, with 43 patients in the NACT + ACT group and 43 patients in the NACT group. Univariate analysis was performed using the Kaplan-Meier method and log-rank test, while Cox regression analysis was used for multivariate analysis.

RESULTS

Multivariate analysis revealed that pathological lymph node status (positive vs negative) (P < .001) and treatment modalities (NACT + ACT vs NACT) (P = .005) were independent prognostic factors. There was a significant difference in long-term survival rates between the NACT + ACT and NACT groups, with 5-year survival rates of 55.8% vs 39.5%, respectively (c2 = 4.270, P = .039). In patients with ypN+ status, the 5-year survival rate was 31.8% for those who received ACT after NACT and surgery, compared to 10.0% for those who did not receive additional ACT (c2 = 6.101, P = .014). The corresponding percentages in patients with ypN- were 81.0% and 65.2%, respectively (c2 = 1.993, P = .158).

CONCLUSION

Adjuvant chemotherapy should be recommended for locally advanced ESCC patients with residual cancer after NACT and surgery, especially for patients with nodal metastases after NACT.

Design and analysis of a compatible exoskeleton rehabilitation robot system based on upper limb movement mechanism

Rehabilitation robots are used to promote structural and functional recovery of the nervous system with repetitive, task-oriented training and have been gradually applied to clinical rehabilitation training. This paper proposes an upper limb exoskeleton rehabilitation robot system that could realize shoulder-elbow-wrist joint rehabilitation training. Firstly, a motion equivalent model was established based on the upper limb movement mechanism, the robot mechanism configuration was designed, and the optimization algorithm and spatial mechanism theory were used to optimize and analyze the structural parameters and human-machine compatibility of the robot, which will guide the design of the robot's model. Then, the robot kinematics were solved, and its maximum motion range, dexterity distribution, and daily motion trajectory were simulated. Finally, a system prototype was built to test the maximum range of robot-assisted human upper limb training by laser tracker, while the pressure of human-machine interaction during training was captured and analyzed by flexible sensors. The results show that the proposed rehabilitation robot could nearly completely cover the range of motion of upper limb joints and meet the needs of trajectory training, and the linear velocity dexterity and angular velocity dexterity in the motion space are maximum 0.55 and 0.89, and the human-machine interaction pressures during the training process are all less than 10 kPa. Besides, this paper also conducted a system evaluation based on the fuzzy comprehensive evaluation model, and the evaluation result was 0.39, with an excellent evaluation grade, it indirectly indicates that the robot's overall performance was good.

Recognizing Predictive Substructures With Subgraph Information Bottleneck

The emergence of Graph Convolutional Network (GCN) has greatly boosted the progress of graph learning. However, two disturbing factors, noise and redundancy in graph data, and lack of interpretation for prediction results, impede further development of GCN. One solution is to recognize a predictive yet compressed subgraph to get rid of the noise and redundancy and obtain the interpretable part of the graph. This setting of subgraph is similar to the information bottleneck (IB) principle, which is less studied on graph-structured data and GCN. Inspired by the IB principle, we propose a novel subgraph information bottleneck (SIB) framework to recognize such subgraphs, named IB-subgraph. However, the intractability of mutual information and the discrete nature of graph data makes the objective of SIB notoriously hard to optimize. To this end, we introduce a bilevel optimization scheme coupled with a mutual information estimator for irregular graphs. Moreover, we propose a continuous relaxation for subgraph selection with a connectivity loss for stabilization. We further theoretically prove the error bound of our estimation scheme for mutual information and the noise-invariant nature of IB-subgraph. Extensive experiments on graph learning and large-scale point cloud tasks demonstrate the superior property of IB-subgraph.

Insights on Using Time-of-Flight Camera for Recovering Cardiac Pulse From Chest Motion in Depth Videos

In this article, we introduce a novel use of depth camera to extract cardiac pulse signal from human chest area, in which the depth information is obtained from a near infrared sensor using time-of-flight technology. We successfully isolate weak chest motion due to heartbeat by processing a sequence of depth images without raising privacy concern. We discuss motion sensitivity in depth video with examples from actuator simulation and human chest motion. Compared to other imaging modalities, the depth image intensity can be directly used for micromotion reconstruction. To deal with the challenges of recovering heartbeat from the chest area, we develop a set of coherent processing techniques to suppress the unwanted motion interference from breathing motion and involuntary body motion and eventually obtain clean cardiac pulse signal. We, thus, derive inter-beat-interval, showing high consistency to the contact photoplethysmography. Additionally, we develop a graphical interpretation of the most and the less pulsatile principal components in eigen space. For validation, we test our method on ten healthy human subjects with different resting heart rates. More importantly, we conduct a set of experiments to study the robustness and weakness of our methods, including extended range, multi-subject, thickness of clothes and generation to other measurement site.

Toward Robust Self-Training Paradigm for Molecular Prediction Tasks

Molecular prediction tasks normally demand a series of professional experiments to label the target molecule, which suffers from the limited labeled data problem. One of the semisupervised learning paradigms, known as self-training, utilizes both labeled and unlabeled data. Specifically, a teacher model is trained using labeled data and produces pseudo labels for unlabeled data. These labeled and pseudo-labeled data are then jointly used to train a student model. However, the pseudo labels generated from the teacher model are generally not sufficiently accurate. Thus, we propose a robust self-training strategy by exploring robust loss function to handle such noisy labels in two paradigms, that is, generic and adaptive. We have conducted experiments on three molecular biology prediction tasks with four backbone models to gradually evaluate the performance of the proposed robust self-training strategy. The results demonstrate that the proposed method enhances prediction performance across all tasks, notably within molecular regression tasks, where there has been an average enhancement of 41.5%. Furthermore, the visualization analysis confirms the superiority of our method. Our proposed robust self-training is a simple yet effective strategy that efficiently improves molecular biology prediction performance. It tackles the labeled data insufficient issue in molecular biology by taking advantage of both labeled and unlabeled data. Moreover, it can be easily embedded with any prediction task, which serves as a universal approach for the bioinformatics community.

Influence of Air Volume and Temperature in the Air Inlet Tunnel on the Characteristics of Dust Movement

In order to clarify the movement characteristics of dust particles in the intake tunnel and improve the underground intake tunnel environment, the main intake tunnel of Wulihou Coal Mine was taken as the engineering background, the COMSOL simulation software was adopted to establish a model, the influence of air volume and temperature on dust movement characteristics was studied, and the critical air volume and particle size of dust at room temperature were determined. The results show that with the ventilation air volume being the same, when the dust reaches the exit of the tunnel, its falling height is positively correlated with the particle size. When the dust particle size is the same, the height of the dust falling is negatively correlated with the ventilation air volume. As the particle size of dust increases, the impact of changes in air volume on its movement decreases and the trajectory of dust movement gradually becomes consistent. The height of dust falling is negatively correlated with the air flow temperature. Therefore, when other factors remain unchanged, dust pollution in the tunnel is relatively severe during the day and the dust concentration in the tunnel is higher in summer. The critical air volume for dust emission in the intake tunnel of Wulihou Coal Mine is 75.84 m3/s, corresponding to a central wind speed of 5.29 m/s in the tunnel. The critical particle size for dust emission is 1.4 μm.

Design and evaluation of a wearable vascular interventional surgical robot system

BACKGROUND

Remote-controlled robotic vascular interventional surgery can reduce radiation exposure to interventional physicians and improve safety. However, inconvenient operation and lack of force feedback limit its application.

MATERIALS AND METHODS

A new wearable robotic system for vascular interventional surgery is designed, which is more flexible in operation. It ensures the safety of surgery through haptic force feedback. The system was evaluated by human vascular models and animal experiments.

RESULTS

The average static error of the system is 0.048 mm when the axial motion is 250 mm and 1.259° when the rotational motion is 400°. The average error of the force feedback is 0.021 N. The results of vascular model experiments and animal experiments demonstrate the feasibility and safety of the system.

CONCLUSIONS

The proposed robotic system can assist physicians in remotely delivering standard catheters or guidewires. The system is more flexible and uses haptic force feedback to ensure surgical safety.

Collaborative and privacy-preserving retired battery sorting for profitable direct recycling via federated machine learning

Unsorted retired batteries with varied cathode materials hinder the adoption of direct recycling due to their cathode-specific nature. The surge in retired batteries necessitates precise sorting for effective direct recycling, but challenges arise from varying operational histories, diverse manufacturers, and data privacy concerns of recycling collaborators (data owners). Here we show, from a unique dataset of 130 lithium-ion batteries spanning 5 cathode materials and 7 manufacturers, a federated machine learning approach can classify these retired batteries without relying on past operational data, safeguarding the data privacy of recycling collaborators. By utilizing the features extracted from the end-of-life charge-discharge cycle, our model exhibits 1% and 3% cathode sorting errors under homogeneous and heterogeneous battery recycling settings respectively, attributed to our innovative Wasserstein-distance voting strategy. Economically, the proposed method underscores the value of precise battery sorting for a prosperous and sustainable recycling industry. This study heralds a new paradigm of using privacy-sensitive data from diverse sources, facilitating collaborative and privacy-respecting decision-making for distributed systems.

The Relationship between Choroidal Thickness and Liver Damage in Simple Auto-immune Hepatitis Patients

BACKGROUND

There was no sufficient clinical evidence on the relationship between auto-immune hepatitis (AIH) and risk of eye illness, except 11 uveitis cases where related AIH is reported currently.

AIM

To determine the relationship between choroidal thickness (ChT) and liver damage in simple AIH patients without ocular symptoms after oral prednisone treatment.

PATIENTS AND METHODS

This prospective observational study included simple AIH patients. The patients' ChT was measured by swept-source (SS)-optical coherence tomography (OCT), and the liver damage was evaluated by alanine aminotransferase (ALT) and aspartate aminotransferase (AST). ChT and liver functions were assessed prior to and after treatment. Then comparisons were made prior to and post treatment. The relationships between biochemical indexes of liver injury and ChT were evaluated after a mean (SD) of 24 (1.28) weeks of regular oral prednisone.

RESULTS

A total of 35 patients (31 females, aged 45.66 ± 11.62 years) were included. After treatment, ChT was significantly increased in all sectors (including the center sector, superior inner sector, inner nasal sector, inferior inner sector, inner temporal sector, superior outer sector, outer nasal sector, inferior outer sector, and outer temporal sector) (all P < 0.001). After treatment, both ALT (51.34 ± 44.16 vs 255.06 ± 107.84, P < 0.001) and AST (38.66 ± 27.12 vs 164.89 ± 85.58, P < 0.001) were significantly decreased. The increase of ChT in all sectors was significantly related to the decrease of ALT and AST (all P < 0.001).

CONCLUSION

The improvement of ChT might reflect the remission of liver damage in simple AIH patients without ocular symptoms during oral prednisone treatment.

Application of pesticide application measures to reduce residue based on the metabolic transfer law of imidacloprid in banana leaves and soil

An investigation of the metabolism and transfer of imidacloprid (IMI) in banana plants and soil was performed using high-resolution mass spectrometry. Results indicated the presence of eight IMI metabolites in soil and leaves that resulted from hydroxylation of the imidazolidine ring, the reduction and loss of nitro groups, and oxidative cleavage of methylene bridges. Six metabolites, including 4/5-hydroxy IMI (4/5-hydroxy), IMI olefin (olefin), and 6-chloronicotinic acid (6-CNA), were detected in the fruits following leaf treatment, while only three were detected after soil treatment. Quantitative analysis showed that the total amount of imidacloprid and its metabolites transferred from leaves to fruits was higher than that transferred from soil to fruits. Therefore, leaf transfer was considered the main means by which IMI and its metabolites transferred to banana fruits. We found that adjuvants tank-mixed with IMI could reduce the total concentration of pesticide transfer from leaves to fruits, especially reducing the amount of metabolites transformed from the reduction and loss of nitro groups and oxidative cleavage of methylene bridges, thus reducing the pesticide residue in fruits and achieving the purpose of reducing the safety risk.

Group membership modulates the hold-up problem: an event-related potentials and oscillations study

This paper investigates the neural mechanism that underlies the effect of group identity on hold-up problems. The behavioral results indicated that the investment rate among members of the in-group was significantly higher than that of the out-group. In comparison to the NoChat treatment, the Chat treatment resulted in significantly lower offers for both in-group and out-group members. The event-related potentials (ERP) results demonstrated the presence of a distinct N2 component in the frontal midline of the brain when investment decisions were made for both in-group and out-group members. During the offer decision-making stage, the P3 peak amplitude was significantly larger when interacting with in-group members compared to the out-group members. The event-related potentials oscillations (ERO) results indicated that when investment decisions were made for in-group members in the NoChat treatment, the beta band (18-28 Hz, 250-350 ms) power was more pronounced than when decisions were made for out-group members. In the NoChat treatment, offer decisions for in-group members yielded a more pronounced difference in beta band (15-20 Hz, 200-300 ms) power when compared to out-group members. Evidence from this study suggests that group identity can reduce the hold-up problem and corroborates the neural basis of group identity.

The impact of mental health literacy intervention on in-service teachers' knowledge attitude and self-efficacy

Mental health literacy (MHL) is an essential component in the process of de-stigmatization, promoting mental health, and supporting people struggling with mental illness. Today, as the number of people suffering from mental illnesses is nearly 450 million people worldwide, the importance of having a teacher-training program that incorporates MHL in its curriculum has become paramount. This study is a quasi-experimental pre-post research that uses the MHL program as an intervention. The sample included 36 (n = 36) high school in-service teachers from 11 schools. The participants took an online training program for 6 weeks. The evaluations were used before and after the training to assess their mental health knowledge, attitude, and self-efficacy. The data were analyzed using the paired sample t-test. The findings revealed a significant level of improvement in relation to the teachers' MHL knowledge, attitude, and self-efficacy after the training. The study findings reveal the effectiveness of MHL training in improving in-service teachers' knowledge, attitude, and self-efficacy in relation to mental health. The study's limitations and future study recommendations are discussed.

Effect of M-type reactive dyes on dyeing properties of twisted bamboo fiber bundles

The dyeing properties of twisted bamboo fiber bundles were studied by using the combination of three primary colors in M-type reactive dyes. The study found that the three dyes of red, yellow and blue have good color rendering in the actual dyeing process, and because the molecular structures of the three are similar, the chemical reactions during the fixation are the same, so the final dyeing rate results are similar, which were 29-32%. Compared with the undyed twisted bamboo fiber bundle, the mechanical properties of the three-color twisted bamboo fiber bundle also changed significantly, and the tensile strength increased by 13.79% on average. The changes of elastic modulus and elongation at break showed that the three-color twisted bamboo fiber bundle had excellent flexibility. In addition, there are significant color differences between the samples of each color. When the color indexes of DB are used as the benchmark, the ΔE* of other dyed samples varies from 40 to 80, and the color of each sample is relatively uniform, without an obvious color flower phenomenon. This indicates that twisted bamboo fiber bundles with richer colors can be prepared by different combinations of three primary dyes, which can improve the ornamental value of bamboo fiber bundles after processing into large blocks and their application potential in home textiles, interior decoration and other fields.

Association between number of dissected lymph nodes and survival in patients undergoing resection for clinical stage IA pure solid lung adenocarcinoma: a retrospective analysis

BACKGROUND

Lymph node dissection is essential for staging of pure solid lung adenocarcinoma and selection of treatment after surgical resection, particularly for stage I disease since the rate of lymph node metastasis can vary from 0 to 23.7%.

METHODS

We retrospectively screened all adult patients (18 years of age or older) who underwent lobectomy for pure solid cT1N0M0 lung adenocarcinoma between January 2015 and December 2017 at our center. Cox proportional hazard regression was used to assess the association between the number of dissected lymph nodes and recurrence-free survival (RFS) and to determine the optimal number of dissected lymph nodes.

RESULTS

The final analysis included 458 patients (age: 60.26 ± 8.07 years; 241 women). RFS increased linearly with an increasing number of dissected lymph nodes at a range between 0 and 9. Kaplan-Meier analysis revealed significantly longer RFS in patients with ≥ 9 vs. <9 dissected lymph nodes. In subgroup analysis, ≥ 9 dissected lymph nodes was not only associated with longer RFS in patients without lymph node metastasis (n = 332) but also in patients with metastasis (n = 126). In multivariate Cox proportional hazard regression, ≥ 9 dissected lymph nodes was independently associated with longer RFS (hazard ratio [HR], 0.43; 95% confidence interval [CI], 0.26 to 0.73; P = 0.002).

CONCLUSIONS

≥9 Dissected lymph nodes was associated with longer RFS; accordingly, we recommend dissecting 9 lymph nodes in patients undergoing lobectomy for stage IA pure solid lung adenocarcinoma.

A Free, Open-Source Toolkit to Produce 3D Bolus in the Clinic

PURPOSE/OBJECTIVE(S)

Tissue-equivalent, tissue-approximating and tissue-replacing bolus materials have been in use for decades in radiotherapy. Most frequently these materials are applied to a patient's skin to bring the highest dose region towards the surface of the skin-which is the location of the target. These materials can be applied at the time of simulation and included in a planning CT scan, or can be added during the planning process and first physically applied at the time of treatment. One of the most widely adopted materials for bolus has been sheets of a commercially available proprietary synthetic gel, which is uniform in thickness, and has some ability to match the curvature of the patient's body. Recently investigators have worked to create boluses using 3D printing technology, including several commercially available offerings. We hypothesized that we could create a bespoke, 3D bolus solution, using a series of open-source and free software products.

MATERIALS/METHODS

For an anthropomorphic phantom, a radiation treatment plan representative of skin cancer treatment was designed, this included a superficial target. The DICOM CT and structure set were imported into 3D Slicer, which is a free, open-source software for visualization, processing, segmentation, and registration. Using 3D Slicer, the bolus structure was saved as an STL file. Meshmixer, a free software for working with triangle meshes, was used to complete a mold design, and the mold parts were then printed using a rigid filament on a 3D printer. The mold parts were glued together, and small spring clamps were used secure the walls to the shells to ensure mold integrity. The mold was then filled with a thinned and degassed silicone. After appropriate curing, demolding was completed by removing the clamps and separating the walls. After QA, the bolus was applied to the anthropomorphic phantom and CTs were taken to compare a commercial sheet bolus with the in-house 3D printed product.

RESULTS

The bolus made via the in-house 3D printing process fit even complicated patient geometries well, and had both an obvious visual/goodness of fit advantage over the commercial sheet bolus and a nuanced dosimetric improvement as the air gaps present in the commercial sheet bolus were not desirable nor reproducible. The overall in-house workflow was efficient, and clinically reasonable (an estimated time of 72 hours was presented to the physician team, but in testing less than 24 hours was needed from export to delivery of the finished product).

CONCLUSION

In this work we explored whether motivated groups and departments could produce dosimetrically accurate and clinically reasonable custom boluses for patients undergoing radiotherapy to a superficial area of the body, using a test case on an anthropomorphic phantom. We found that this was absolutely achievable and could be implemented with no funds spent on software or licenses. Provided that a 3D printer, filament and silicone are available, any thoughtful practice can join the bespoke-bolus-club.

scMHNN: a novel hypergraph neural network for integrative analysis of single-cell epigenomic, transcriptomic and proteomic data

Technological advances have now made it possible to simultaneously profile the changes of epigenomic, transcriptomic and proteomic at the single cell level, allowing a more unified view of cellular phenotypes and heterogeneities. However, current computational tools for single-cell multi-omics data integration are mainly tailored for bi-modality data, so new tools are urgently needed to integrate tri-modality data with complex associations. To this end, we develop scMHNN to integrate single-cell multi-omics data based on hypergraph neural network. After modeling the complex data associations among various modalities, scMHNN performs message passing process on the multi-omics hypergraph, which can capture the high-order data relationships and integrate the multiple heterogeneous features. Followingly, scMHNN learns discriminative cell representation via a dual-contrastive loss in self-supervised manner. Based on the pretrained hypergraph encoder, we further introduce the pre-training and fine-tuning paradigm, which allows more accurate cell-type annotation with only a small number of labeled cells as reference. Benchmarking results on real and simulated single-cell tri-modality datasets indicate that scMHNN outperforms other competing methods on both cell clustering and cell-type annotation tasks. In addition, we also demonstrate scMHNN facilitates various downstream tasks, such as cell marker detection and enrichment analysis.

Structure-Aware DropEdge Toward Deep Graph Convolutional Networks

It has been discovered that graph convolutional networks (GCNs) encounter a remarkable drop in performance when multiple layers are piled up. The main factor that accounts for why deep GCNs fail lies in oversmoothing, which isolates the network output from the input with the increase of network depth, weakening expressivity and trainability. In this article, we start by investigating refined measures upon DropEdge-an existing simple yet effective technique to relieve oversmoothing. We term our method as DropEdge ++ for its two structure-aware samplers in contrast to DropEdge: layer-dependent (LD) sampler and feature-dependent (FD) sampler. Regarding the LD sampler, we interestingly find that increasingly sampling edges from the bottom layer yields superior performance than the decreasing counterpart as well as DropEdge. We theoretically reveal this phenomenon with mean-edge-number (MEN), a metric closely related to oversmoothing. For the FD sampler, we associate the edge sampling probability with the feature similarity of node pairs and prove that it further correlates the convergence subspace of the output layer with the input features. Extensive experiments on several node classification benchmarks, including both full-and semi-supervised tasks, illustrate the efficacy of DropEdge ++ and its compatibility with a variety of backbones by achieving generally better performance over DropEdge and the no-drop version.

RNA-Seq Analysis Reveals an Essential Role of the cGMP-PKG-MAPK Pathways in Retinal Degeneration Caused by Cep250 Deficiency

Usher syndrome (USH) is characterised by degenerative vision loss known as retinitis pigmentosa (RP), sensorineural hearing loss, and vestibular dysfunction. RP can cause degeneration and the loss of rod and cone photoreceptors, leading to structural and functional changes in the retina. Cep250 is a candidate gene for atypical Usher syndrome, and this study describes the development of a Cep250 KO mouse model to investigate its pathogenesis. OCT and ERG were applied in Cep250 and WT mice at P90 and P180 to access the general structure and function of the retina. After recording the ERG responses and OCT images at P90 and P180, the cone and rod photoreceptors were visualised using an immunofluorescent stain. TUNEL assays were applied to observe the apoptosis in Cep250 and WT mice retinas. The total RNA was extracted from the retinas and executed for RNA sequencing at P90. Compared with WT mice, the thickness of the ONL, IS/OS, and whole retina of Cep250 mice was significantly reduced. The a-wave and b-wave amplitude of Cep250 mice in scotopic and photopic ERG were lower, especially the a-wave. According to the immunostaining and TUNEL stain results, the photoreceptors in the Cep250 retinas were also reduced. An RNA-seq analysis showed that 149 genes were upregulated and another 149 genes were downregulated in Cep250 KO retinas compared with WT mice retinas. A KEGG enrichment analysis indicated that cGMP-PKG signalling pathways, MAPK signalling pathways, edn2-fgf2 axis pathways, and thyroid hormone synthesis were upregulated, whereas protein processing in the endoplasmic reticulum was downregulated in Cep250 KO eyes. Cep250 KO mice experience a late-stage retinal degeneration that manifests as the atypical USH phenotype. The dysregulation of the cGMP-PKG-MAPK pathways may contribute to the pathogenesis of cilia-related retinal degeneration.

Semi-Supervised Hierarchical Graph Classification

Node classification and graph classification are two graph learning problems that predict the class label of a node and the class label of a graph respectively. A node of a graph usually represents a real-world entity, e.g., a user in a social network, or a document in a document citation network. In this work, we consider a more challenging but practically useful setting, in which a node itself is a graph instance. This leads to a hierarchical graph perspective which arises in many domains such as social network, biological network and document collection. We study the node classification problem in the hierarchical graph where a "node" is a graph instance. As labels are usually limited, we design a novel semi-supervised solution named SEAL-CI. SEAL-CI adopts an iterative framework that takes turns to update two modules, one working at the graph instance level and the other at the hierarchical graph level. To enforce a consistency among different levels of hierarchical graph, we propose the Hierarchical Graph Mutual Information (HGMI) and further present a way to compute HGMI with theoretical guarantee. We demonstrate the effectiveness of this hierarchical graph modeling and the proposed SEAL-CI method on text and social network data.

Enhancer variants on chromosome 2p14 regulating SPRED2 and ACTR2 act as a signal amplifier to protect against rheumatoid arthritis

Genome-wide association studies (GWASs) have repeatedly reported multiple non-coding single-nucleotide polymorphisms (SNPs) at 2p14 associated with rheumatoid arthritis (RA), but their functional roles in the pathological mechanisms of RA remain to be explored. In this study, we integrated a series of bioinformatics and functional experiments and identified three intronic RA SNPs (rs1876518, rs268131, and rs2576923) within active enhancers that can regulate the expression of SPRED2 directly. At the same time, SPRED2 and ACTR2 influence each other as a positive feedback signal amplifier to strengthen the protective role in RA by inhibiting the migration and invasion of rheumatoid fibroblast-like synoviocytes (FLSs). In particular, the transcription factor CEBPB preferentially binds to the rs1876518-T allele to increase the expression of SPRED2 in FLSs. Our findings decipher the molecular mechanisms behind the GWAS signals at 2p14 for RA and emphasize SPRED2 as a potential candidate gene for RA, providing a potential target and direction for precise treatment of RA.

Contactless Stethoscope Enabled by Radar Technology

Contactless vital sign measurement technologies have the potential to greatly improve patient experiences and practitioner safety while creating the opportunity for comfortable continuous monitoring. We introduce a contactless alternative for measuring human heart sounds. We leverage millimeter wave frequency-modulated continuous wave radar and multi-input multi-output beamforming techniques to capture fine skin vibrations that result from the cardiac movements that cause heart sounds. We discuss contact-based heart sound measurement techniques and directly compare the radar heart sound technique with these contact-based approaches. We present experimental cases to test the strengths and limitations of both the contact-based measurement techniques and the contactless radar measurement. We demonstrate that the radar measurement technique is a viable and potentially superior method for capturing human heart sounds in many practical settings.

Spatiotemporal impacts of climate change and human activities on water resources and ecological sensitivity in the Mekong subregion in Cambodia

Water resources in the Mekong subregion in Cambodia (MSC) have experienced dramatic changes in past decades, threatening regional ecosystem quality and sustainable development. Thus, it is important to explore the spatiotemporal impacts of climate change and human activities on water resources and ecological sensitivity. This study proposed an effective framework including spatiotemporal analysis of land use/cover change (LUCC) and ecological sensitivity assessment by combining remote sensing (RS) and geographic information system/science (GIS). An optimized feature space and a machine learning classification algorithm were constructed to extract four typical land cover types in the MSC from 1990 to 2020. An ecological sensitivity evaluation system, including four sub-sensitivities calculated by twelve indicators, was then constructed. The results suggest that severe shrinkage of water resources occurred before 2006, decreasing by 21.68%. The correlation between water resources and climate conditions displays a high to low level as human activity becomes involved. A significant spatiotemporal evolutionary pattern of ecological sensitivity was observed under the impact of external interference. Generally, the largest proportion of MSC belongs to the lightly sensitive level, which is mainly concentrated in the lower reaches, with an average of 33.93%. The highly sensitive area with a significant value in ecological protection has a slightly downward trend from 23.72 in 1990 to 22.55% in 2020.

Investigation on Mechanical and Microstructure Properties of Silt Improved by Titanium Gypsum-Based Stabilizer

Silt in the Yellow River alluvial plain is widely spread, but its uniform particle size and high roundness make it unsuitable as a subgrade filling material, while titanium gypsum (TG) is an industrial solid waste in Shandong Province, not only occupying land resources but also causing water and air pollution. In order to improve the engineering performance of silt, reduce the pollution of solid waste titanium gypsum to the environment and reduce the engineering cost, considering the engineering characteristics of titanium gypsum, it was combined in specific amounts with cement and lime to create a titanium gypsum-based stabilizer (TS) in this study. The effect of curing conditions and TS content on silt improvement was studied through laboratory experiments. The mechanical properties of the stabilized silt were investigated by unconfined compression test (UCT), and the mineral composition and pore structure were analyzed by scanning electron microscopy (SEM) test, X-ray diffraction (XRD) test, and mercury injection pore (MIP) test. The test results show that TS could effectively improve the unconfined compressive strength of silt, and the strength of stabilized silt gradually increases with the curing period and TS content. In terms of the porosity, it decreases with the increase of the curing period and TS content. From the microstructure perspective, this is mainly due to the formation of ettringite and C-S-H during the stabilization of silt by TS.

Methylmalonic acid levels in serum, exosomes, and urine and its association with cblC type methylmalonic acidemia-induced cognitive impairment

Background

The cblC type methylmalonic acidemia is the most common methylmalonic acidemia (MMA) in China. The biochemical characteristics of this disease include elevated methylmalonic acid and homocysteine (HCY), increased propionylcarnitine (C3), decreased free carnitine (C0). In this study, we aimed to clarify the roles of these biomarkers in cblC-MMA induced cognitive impairment and evaluate the capacity of methylmalonic acid in different fluids or exosomes to distinguish cblC-MMA induced cognitive impairment.

Methods

15 non-inherited hyperhomocysteinemia (HHcy) patients, 42 cblC-MMA patients and 57 age- and sex-matched healthy children were recruited in this study. The levels of HCY were detected by an automatic immune analyzer. The levels of acylcarnitines and methylmalonic acid were detected by tandem mass spectrometer.

Results

The main findings were all biomarkers as HCY, acylcarnitines and methylmalonic acid had capacities for distinguishing patients with cblC-MMA induced cognitive impairment from healthy children. The methylmalonic acid in different fluids or exosomes had good performances for distinguishing patients with cblC-MMA induced cognitive impairment from HHcy patients. The methylmalonic acid in serum exosomes and neuronal-derived exosomes were able to distinguishing cblC-MMA patients with cognitive impairment from patients without cognitive impairment. The methylmalonic acid in neuronal-derived exosomes might be helpful to evaluate the severity of cblC-MMA induced cognitive impairment.

Discussion

Methylmalonic acid levels in serum exosomes, especially in serum neuronal-derived exosomes, serve as potential biomarkers for distinguishing cblC-MMA induced cognitive impairment.

Lineage-specific rearrangement of chromatin loops and epigenomic features during adipocytes and osteoblasts commitment

Human mesenchymal stem cells (hMSCs) can be differentiated into adipocytes and osteoblasts. The processes are driven by the rewiring of chromatin architectures and transcriptomic/epigenomic changes. Here, we induced hMSCs to adipogenic and osteogenic differentiation, and performed 2 kb resolution Hi-C experiments for chromatin loops detection. We also generated matched RNA-seq, ChIP-seq and ATAC-seq data for integrative analysis. After comprehensively comparing adipogenesis and osteogenesis, we quantitatively identified lineage-specific loops and screened out lineage-specific enhancers and open chromatin. We reveal that lineage-specific loops can activate gene expression and facilitate cell commitment through combining enhancers and accessible chromatin in a lineage-specific manner. We finally proposed loop-mediated regulatory networks and identified the controlling factors for adipocytes and osteoblasts determination. Functional experiments validated the lineage-specific regulation networks towards IRS2 and RUNX2 that are associated with adipogenesis and osteogenesis, respectively. These results are expected to help better understand the chromatin conformation determinants of hMSCs fate commitment.

A Rare Presentation of Thoracic Intramedullary Chordoma with Adjacent Bone Involvement: A Case Report and Review of the Literature

Introduction/Objective

Chordoma originates from remnants of the embryonal notochord, and arise in bones anywhere along the spine and skull base. The most common location was thought to be the sacrum, followed by the clivus, and to a much lesser extent the rest of the spine. However, some studies have suggested an equal distribution among the skull base (32%), mobile spine (32.8%), and sacro-coccygeal bones (29.2%). Here we report a case of chordoma involving the thoracic spine. at the level of T2.

Methods/Case Report

A 63-year-old male with no significant past medical history who presented with 5-6 months of intermittent, bilateral lower extremity weakness and numbness in the trunk and lower extremities. MRI of the thoracic spine demonstrated a contrast enhancing mass at T2 vertebral level with spinal cord compression and adjacent bone destruction. T1-3 laminectomy with debulking of the tumor was performed. Microscopically, the tumor cells have a lobulated architecture and are composed of epithelioid cells arranged in cords, clusters or nests, embedded in a myxoid mucinous matrix. The epithelioid cells have a variably vacuolated cytoplasm ("physaliphorous" cells). The epithelioid cells are positive for CK AE1/3, Cam5.2, EMA and Brachyury (nuclear stain), and S100 (focal). These findings support a diagnosis of chordoma.

Results (if a Case Study enter NA)

N/A.

Conclusion

The most important and difficult differential diagnosis of chordoma is with well-differentiated chondrosarcoma. Although both chordomas and chondrosarcomas express S100, chondrosarcomas do not express cytokeratins, EMA or brachyury. Chordomas have an aggressive clinical course and poor outcome with local extension, recurrence and even metastasis. The treatment is en block surgical resection with adjuvant radiotherapy. The extent of the initial surgical resection is the most significant prognostic factor.

Sudden Cardiac Death due to Giant Cell Myocarditis in a 37-Year-Old Female: Case Report

Introduction/Objective

Giant cell myocarditis (GCM) is a rare and aggressive inflammatory process that targets the myocardium and is often rapidly fatal. Most cases have been reported in young to middle aged adults with a slight male predominance. The etiology of this disease is largely unknown, however there is an association with multiple autoimmune disorders. Most patients present with rapidly progressive or fulminant heart failure, arrhythmias, heart block or sudden cardiac death. The incidence of GCM ranges from 0.007% to 0.051% and most cases are confirmed on autopsy examination. Here we present a case of GCM in a young female diagnosed on autopsy.

Methods/Case Report

A 37-year-old female presented with shortness of breath and chest pain for a 5-day duration. She had no past medical history and was otherwise well before the onset of symptoms. On admission, she was noted to have clinical findings consistent with heart failure and progressed to cardiogenic shock with ventricular arrhythmias within hours. Despite high dose corticosteroids and other supportive therapy, the patient remained in refractory, vasodilatory shock, in cardiac standstill and eventually died. Findings on autopsy included mild ventricular hypertrophy, petechial hemorrhages through out the atria and ventricles, as well as pulmonary edema and congestion. Microscopic examination showed a diffuse infiltrating pattern of inflammation within the heart, that was composed predominantly of lymphocytes, giant cells, scattered neutrophils, and eosinophils. The lymphocytes were CD3, CD4, CD8 positive and CD20 negative, consistent with a T-cell infiltrate. Gram and GMS stains were negative for bacterial and fungal microorganisms. Focal areas of necrosis without granuloma formation are also noted ruling out sarcoidosis.

Results (if a Case Study enter NA)

NA.

Conclusion

This case demonstrates a rapid manifestation of GCM in a previously healthy individual. GCM must be considered in the differential diagnosis when a young patient presents with cardiac symptoms. Early diagnosis and initiation of targeted therapy is critical for GCM patients' survival.

Impact of COVID 19 Pandemic on Radiation Oncology Care Path Time

Purpose/Objective(s)

We seek to investigate the impact of the COVID19 pandemic on the radiation oncology care path timeline. We hypothesized that the COVID19 pandemic would result in increased time to complete the care path from simulation to start of radiation.

Materials/Methods

Care path data of patients treated at a single institution were included for 3 epochs: Pre COVID (July 2019 – March 2020), C1 [PreV] (April 2020 – Dec 2020), and C2 [PostV] (Jan 2021 – September 2021). The following tasks in the care path were evaluated: simulation start time, contours completed, physician review, plan finalization, IMRT/IMPT QA, and radiation treatment start. Time stamps of completion of tasks in the radiation treatment planning care pathway were extrapolated and used to create 9-time intervals. Demographic, tumor, and treatment characteristics were gathered to stratify patients. Patients with care path times greater than 25 days were excluded from the analysis. Numerical variables are summarized using mean and standard deviation while categorical data is summarized using counts and percentages. The ANOVA test is used to compare means in RT planning times between time epochs.

Results

A total of 3557 patients were included in the study analysis: 1095 PC, 1105 C1, and 1357 C2. There were improvements in multiple car path intervals following the start of the pandemic. Across epochs, the mean time from simulation to RT Start was 10.5 days (10.9 PC, 10.3C1, and 10.3 C2) (p =0.008). The mean time from simulation to plan finalization was 5.5 days (5.6 PC, 5.7 C1, 5.2 C2, p =0.008). The mean time for plan review to start time was 5.5 days (5.7 PC, 5.2 C1, and 5.5 C2) (p=0.010). Contour completion to radiation start time was significantly shorter after the pandemic (8.8 PC vs. 8.4 C1 vs. 8.4 C2, p =0.046). The mean time from physician review to IMRT/IMPT QA was 1.7 days (1.8 PC, 1.7 C1, 1.5 C2, p = 0.007). Physician review to RT start took a mean time of 5.5 days (5.7 PC, 5.2 C1, 5.5 C2, p = 0.010). Plan finalization to IMRT/IMPT QA took a mean time of 1.1 days (1.2 PC, 1.1 C1, 1.0 C2, p = 0.01). IMRT/QA to RT start took a mean time of 4.4 days (4.6 PC, 4.0 C1, and 4.5 C2, p =0.002). The mean time from plan finalization to RT start was 5.0 days (5.2 PC, 4.6 C1, 5.0 C2, p = 0.004). There were no significant differences in simulation to contour time (p =0.181) and physician review completion to plan finalization (p =0.223)

Conclusion

The COVID19 pandemic has been a disruptive force in the management of patients in Radiation Oncology. However, we demonstrate improved efficiency in care path completion as we entered into the pandemic. Further investigation into care path efficiency is important for provider and patient satisfaction.

Comparison of complications and long-term survival after minimally invasive esophagectomy versus open esophagectomy in patients with esophageal cancer and chronic obstructive pulmonary disease

Objective

To compare the complications and long-term survival of esophageal cancer patients with chronic obstructive pulmonary disease (COPD) after minimally invasive esophagectomy (MIE) versus open esophagectomy (OE) using propensity score matching (PSM).

Methods

Esophageal cancer patients who underwent esophagectomy at the Thoracic Surgery Department of the First Affiliated Hospital of Hebei North University from January 2010 to December 2018 were retrospectively enrolled. The incidence of postoperative complications and prognosis of the MIE (n = 132) and OE (n = 138) groups were compared. To reduce bias, 1:1 PSM was adopted for the analysis.

Results

The median disease-free survival (DFS) of the MIE and OE groups were 24 months and 26 months, respectively, and neither group reached median survival. There was no significant difference between the two groups in terms of 3-year DFS and overall survival (OS). The stratification of the patients on the basis of the percentage of estimated forced expiratory volume in the first second (%FEV1) did not result in significant differences in the survival rates. A total of 42 patients (50%) in the MIE group and 55 patients (65.48%) in the OE group experienced complications, and the difference was statistically significant (OR=0.527, 95% CI: 0.283–0.981, P=0.042). The incidence of acute COPD exacerbation (OR=0.213, 95% OR, CI: 0.068–0.666, P=0.004) and pulmonary atelectasis requiring bronchoscopic aspiration (OR=0.232, 95% OR, CI: 0.082–0.659, P=0.004) were significantly higher in the OE versus the MIE group. In addition, the distribution of the various grades of complications also differed significantly between the two groups (P=0.016). While the incidence of minor complications (≤Grade II) was similar in both groups (P=0.503), that of severe complications (≥Grade III) was markedly higher in the OE group (P=0.002) and the Grade-IIIa complications were predominant (P=0.001). The severity of complications was correlated with the postoperative duration of hospital stay in both groups (r=0.187, P=0.015). No significant difference was observed in the incidence of minor complications (≤Grade II) between the two groups following stratification on the basis of %FEV1, whereas severe complications were more frequent in the OE group among patients with %FEV1 between 60% and 70% (P=0.001<0.05).

Conclusion

There was no significant difference in the postoperative DFS and OS of esophageal cancer patients with COPD after undergoing MIE or OE. However, MIE significantly reduced the incidence of severe postoperative complications among patients with %FEV1 between 60% and 70%.

Graph Convolutional Module for Temporal Action Localization in Videos

Temporal action localization, which requires a machine to recognize the location as well as the category of action instances in videos, has long been researched in computer vision. The main challenge of temporal action localization lies in that videos are usually long and untrimmed with diverse action contents involved. Existing state-of-the-art action localization methods divide each video into multiple action units (i.e., proposals in two-stage methods and segments in one-stage methods) and then perform action recognition/regression on each of them individually, without explicitly exploiting their relations during learning. In this paper, we claim that the relations between action units play an important role in action localization, and a more powerful action detector should not only capture the local content of each action unit but also allow a wider field of view on the context related to it. To this end, we propose a general graph convolutional module (GCM) that can be easily plugged into existing action localization methods, including two-stage and one-stage paradigms. To be specific, we first construct a graph, where each action unit is represented as a node and their relations between two action units as an edge. Here, we use two types of relations, one for capturing the temporal connections between different action units, and the other one for characterizing their semantic relationship. Particularly for the temporal connections in two-stage methods, we further explore two different kinds of edges, one connecting the overlapping action units and the other one connecting surrounding but disjointed units. Upon the graph we built, we then apply graph convolutional networks (GCNs) to model the relations among different action units, which is able to learn more informative representations to enhance action localization. Experimental results show that our GCM consistently improves the performance of existing action localization methods, including two-stage methods (e.g., CBR [15] and R-C3D [47]) and one-stage methods (e.g., D-SSAD [22]), verifying the generality and effectiveness of our GCM. Moreover, with the aid of GCM, our approach significantly outperforms the state-of-the-art on THUMOS14 (50.9 percent versus 42.8 percent). Augmentation experiments on ActivityNet also verify the efficacy of modeling the relationships between action units. The source code and the pre-trained models are available at https://github.com/Alvin-Zeng/GCM.

High-Stretchability, Ultralow-Hysteresis ConductingPolymer Hydrogel Strain Sensors for Soft Machines

Highly stretchable strain sensors based on conducting polymer hydrogel are rapidly emerging as a promising candidate toward diverse wearable skins and sensing devices for soft machines. However, due to the intrinsic limitations of low stretchability and large hysteresis, existing strain sensors cannot fully exploit their potential when used in wearable or robotic systems. Here, a conducting polymer hydrogel strain sensor exhibiting both ultimate strain (300%) and negligible hysteresis (<1.5%) is presented. This is achieved through a unique microphase semiseparated network design by compositing poly(3,4-ethylenedioxythiophene):polystyrene sulfonate (PEDOT:PSS) nanofibers with poly(vinyl alcohol) (PVA) and facile fabrication by combining 3D printing and successive freeze-thawing. The overall superior performances of the strain sensor including stretchability, linearity, cyclic stability, and robustness against mechanical twisting and pressing are systematically characterized. The integration and application of such strain sensor with electronic skins are further demonstrated to measure various physiological signals, identify hand gestures, enable a soft gripper for objection recognition, and remote control of an industrial robot. This work may offer both promising conducting polymer hydrogels with enhanced sensing functionalities and technical platforms toward stretchable electronic skins and intelligent robotic systems.

Blood-brain barrier penetration prediction enhanced by uncertainty estimation

Blood–brain barrier is a pivotal factor to be considered in the process of central nervous system (CNS) drug development, and it is of great significance to rapidly explore the blood–brain barrier permeability (BBBp) of compounds in silico in early drug discovery process. Here, we focus on whether and how uncertainty estimation methods improve in silico BBBp models. We briefly surveyed the current state of in silico BBBp prediction and uncertainty estimation methods of deep learning models, and curated an independent dataset to determine the reliability of the state-of-the-art algorithms. The results exhibit that, despite the comparable performance on BBBp prediction between graph neural networks-based deep learning models and conventional physicochemical-based machine learning models, the GROVER-BBBp model shows greatly improvement when using uncertainty estimations. In particular, the strategy combined Entropy and MC-dropout can increase the accuracy of distinguishing BBB + from BBB − to above 99% by extracting predictions with high confidence level (uncertainty score < 0.1). Case studies on preclinical/clinical drugs for Alzheimer’ s disease and marketed antitumor drugs that verified by literature proved the application value of uncertainty estimation enhanced BBBp prediction model, that may facilitate the drug discovery in the field of CNS diseases and metastatic brain tumors.

Correlation between the Expression of VEGF and Ki67 and Lymph Node Metastasis in Non-small-Cell Lung Cancer: A Systematic Review and Meta-Analysis

Background

Lymph node metastasis is the most common and important way of metastasis in NSCLC and is also the most important factor affecting lung cancer stage and prognosis. It is very important to analyze the relationship between the expression of vascular endothelial growth factor (VEGF) and Ki67 and lymph node metastasis (LNM) in non-small-cell lung cancer (NSCLC).

Methods

We searched the PubMed, EMBASE, and Cochrane Library and conducted meta-analyses using the R meta-package. Relative risk (RR) with a 95% confidence interval (95% CI) was the main indicator.

Results

Totally, 18 studies were considered eligible, with 4521 patients, including 1518 LNM-positive patients and 3033 LNM-negative patients. The incidence of LNM in Ki67-negative patients was lower than that in Ki67-positive patients (RR = 0.66, 95% CI: 0.44, 0.98). The incidence of LNM in VEGF-A-negative patients was lower than that in VEGF-A-positive patients (RR = 0.64, 95% CI: 0.49, 0.83). The incidence of LNM in VEGF-C negative patients was lower than that in VEGF-C positive patients (RR = 0.68, 95% CI: 0.53, 0.88). The incidence of LNM in VEGF-D negative and positive patients were of no significant differences (RR = 0.84, 95% CI: 0.61, 1.14).

Conclusion

The high expression of Ki67, VEGF-A, and VEGF-C significantly increases the risk of lymph node metastasis in NSCLC, while the VEGF-D expression has no correlation with lymph node metastasis. The expression levels of Ki67, VEGF-A, and VEGF-C show a good potential for lymph node metastasis prediction.

An Osteoporosis Susceptibility Allele at 11p15 Regulates SOX6 Expression by Modulating TCF4 Chromatin Binding

Osteoporosis is an age-related complex disease clinically diagnosed with bone mineral density (BMD). Although several genomewide association studies (GWASs) have discovered multiple noncoding genetic variants at 11p15 influencing osteoporosis risk, the functional mechanisms of these variants remain unknown. Through integrating bioinformatics and functional experiments, a potential functional single-nucleotide polymorphism (SNP; rs1440702) located in an enhancer element was identified and the A allele of rs1440702 acted as an allelic specificities enhancer to increase its distal target gene SOX6 (~600 Kb upstream) expression, which plays a key role in bone formation. We also validated this long-range regulation via conducting chromosome conformation capture (3C) assay. Furthermore, we demonstrated that SNP rs1440702 with a risk allele (rs1440702-A) could increase the activity of the enhancer element by altering the binding affinity of the transcription factor TCF4, resulting in the upregulation expression of SOX6 gene. Collectively, our integrated analyses revealed how the noncoding genetic variants (rs1440702) affect osteoporosis predisposition via long-range gene regulatory mechanisms and identified its target gene SOX6 for downstream biomarker and drug development. © 2022 American Society for Bone and Mineral Research (ASBMR).

ExsgRNA: reduce off-target efficiency by on-target mismatched sgRNA

Clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 gene editing technology has been widely used to facilitate efficient genome editing. Current popular sgRNA design tools only consider the sgRNA perfectly matched to the target site and provide the results without any on-target mismatch. We suppose taking on-target gRNA-DNA mismatches into consideration might provide better sgRNA with similar binding activity and reduced off-target sites. Here, we trained a seq2seq-attention model with feedback-loop architecture, to automatically generate sgRNAs with on-target mismatches. Dual-luciferase reporter experiment showed that multiple sgRNAs with three mismatches could achieve the 80% of the relative activity of the perfect matched sgRNA. Meanwhile, it could reduce the number of off-target sites using sgRNAs with on-target mismatches. Finally, we provided a freely accessible web server sgRNA design tool named ExsgRNA. Users could submit their target sequence to this server and get optimal sgRNAs with less off-targets and similar on-target activity compared with the perfect-matched sgRNA.

Chasing the Tail in Monocular 3D Human Reconstruction With Prototype Memory

Deep neural networks have achieved remarkable progress in single-image 3D human reconstruction. However, existing methods still fall short in predicting rare poses. The reason is that most of the current models perform regression based on a single human prototype, which is similar to common poses while far from the rare poses. In this work, we 1) identify and analyze this learning obstacle and 2) propose a prototype memory-augmented network, PM-Net, that effectively improves performances of predicting rare poses. The core of our framework is a memory module that learns and stores a set of 3D human prototypes capturing local distributions for either common poses or rare poses. With this formulation, the regression starts from a better initialization, which is relatively easier to converge. Extensive experiments on several widely employed datasets demonstrate the proposed framework's effectiveness compared to other state-of-the-art methods. Notably, our approach significantly improves the models' performances on rare poses while generating comparable results on other samples.

Cross-dependent graph neural networks for molecular property prediction

MOTIVATION

The crux of molecular property prediction is to generate meaningful representations of the molecules. One promising route is to exploit the molecular graph structure through graph neural networks (GNNs). Both atoms and bonds significantly affect the chemical properties of a molecule, so an expressive model ought to exploit both node (atom) and edge (bond) information simultaneously. Inspired by this observation, we explore the multi-view modeling with GNN (MVGNN) to form a novel paralleled framework, which considers both atoms and bonds equally important when learning molecular representations. In specific, one view is atom-central and the other view is bond-central, then the two views are circulated via specifically designed components to enable more accurate predictions. To further enhance the expressive power of MVGNN, we propose a cross-dependent message-passing scheme to enhance information communication of different views. The overall framework is termed as CD-MVGNN.

RESULTS

We theoretically justify the expressiveness of the proposed model in terms of distinguishing non-isomorphism graphs. Extensive experiments demonstrate that CD-MVGNN achieves remarkably superior performance over the state-of-the-art models on various challenging benchmarks. Meanwhile, visualization results of the node importance are consistent with prior knowledge, which confirms the interpretability power of CD-MVGNN.

AVAILABILITY AND IMPLEMENTATION

The code and data underlying this work are available in GitHub at https://github.com/uta-smile/CD-MVGNN.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.