Search

Additional file 1: Figure S1. SS-31 does not affect the histological morphology and motor function in non-SCI mice. (A) Photographs of spinal cord sections in the Sham and Sham+SS-31 groups stained with antibody MAP2 (green) (scale bar = 25 μm). (B) MAP2 optical density within uninjured spinal cord on day 28. (C) Photographs of spinal cord sections in the respective groups stained with antibody Syn (green)/NeuN (red) (scale bar = 20 μm). (D) Relevant quantitative results for motor neuron-contacting synapse amounts on day 28 after sham surgery. (E) Photographs of mouse footprints on day 28 following sham surgery. Blue: forepaw print; Red: hind paw print. (F, G) Toe dragging (%) and stride length (cm) analyses of mice at 28 days after sham surgery. (H) Longitudinal spinal cord sections from the groups at 28 days after sham surgery were examined via HE dyeing and Masson dyeing (scale bar = 1000 μm). (I) Basso mouse scale (BMS) for the indicated groups at days 0, 1, 3, 7, 14, 21, and 28. The data are shown as the mean ± SEM. n = 5. *P < 0.05, **P < 0.01. ns indicates no significance. Figure S2. SS-31 attenuates pyroptosis in microglia after SCI. (A) Typical images of immunofluorescence staining for Caspase-1 and microglia colocalization and in the spinal cords of the Sham, SCI, and SCI+SS-31 groups (scale bar= 25μm) (B) Quantified Caspase-1 immunofluorescence data are presented on the right. (C) Typical images of immunofluorescence staining for NLRP3 and microglia colocalization and in the spinal cords of the Sham, SCI, and SCI+SS-31 groups (scale bar= 25μm) (D) Quantified NLRP3 immunofluorescence data are presented on the right. The white arrows indicate Caspase-1 or NLRP3-positive microglia. The data are shown as the mean ± SEM. n = 5. *P < 0.05, **P < 0.01. ns indicates no significance. Figure S3. CQ treatment does not affect pyroptosis in non-SCI mice. (A-D) Representative double immunofluorescence images for Caspase-1/NeuN and NLRP3/NeuN in spinal cord from Sham and Sham+CQ groups at 3 dpi (scale bar= 25μm). The quantitative integrated densities of Caspase-1 and NLRP3 in each neuron are shown on the graph. (E) Typical images of WB results of Caspase-1, GSDMD-N, NLRP3, NLRP1, ASC, IL-1β, IL-18 in the spinal cord. GAPDH was utilized as a loading control. (F) Quantification results of the pyroptosis-related protein levels from (E) in both groups. The data are shown as the mean ± SEM. ns indicates no significance. Figure S4. SS-31’s functional recovery effect is inhibited by CQ. (A) Photographs of spinal cord sections in the SCI, SCI+SS-31, SCI+SS-31+CQ groups stained with antibody MAP2 (green) (scale bar = 25 μm). Photographs of spinal cord sections in the respective groups stained with Syn (green)/NeuN (red) (scale bar = 20 μm). (B) Relevant quantitative results for motor neuron-contacting synapse amounts on day 28 after SCI. (C) MAP2 optical density within a spinal cord subjected to injury on day 28. (D) Longitudinal spinal cord sections at 28 dpi after SCI stained with HE dyeing and Masson dyeing (scale bar = 1000 μm). (E) The lesion area in injured spinal cord measured from Masson staining. (F) Photographs of mouse footprints on day 28 following spinal cord injury. Blue: forepaw print; Red: hind paw print. (G, H) Toe dragging (%) and stride length (cm) analyses of mice at 28 dpi after SCI. (I, J) Inclined plane test and Basso mouse scale (BMS) for the indicated groups at days 0, 1, 3, 7, 14, 21, and 28. The data are shown as the mean ± SEM. n = 5. *P < 0.05, **P < 0.01. ns indicates no significance. Figure S5. SS-31 affects the p-cPLA2 expression in glial cells. (A-B) Typical images of immunofluorescence staining for p-cPLA2 and microglia colocalization in the spinal cords of the Sham, SCI, and SCI+SS-31 groups. The white arrow indicates the p-cPLA2 positive microglia. The ratio of p-cPLA2 positive cells to the total is shown on the right. (C-D) Typical images of immunofluorescence staining for p-cPLA2 and oligodendrocyte colocalization in the spinal cords of three groups. The white arrow indicates the p-cPLA2 positive oligodendrocyte. The ratio of p-cPLA2 positive cells to the total is shown on the right. (E-F) Typical images of immunofluorescence staining for p-cPLA2 and astrocyte colocalization in the spinal cords of three groups. The white arrow indicates the p-cPLA2 positive astrocyte. The ratio of p-cPLA2 positive cells to the total is shown on the right. The data are shown as the mean ± SEM. n = 5. *P < 0.05, **P < 0.01. ns indicates no significance. Figure S6. SS-31 inhibits the activity of cPLA2 and protect the lysosomal enzymes activity in lysosomes. (A) ELISA results showing the activity of cPLA2 in the spinal cord from Sham, SCI, and SCI+SS-31 groups on postoperative day 3. (B-D) ELISA results of the activity of lysosomal enzymes CTSD and NAGLU in lysosomal (B, C) and cytosolic (D, E) fractions extracted from the spinal cord of Sham, SCI, and SCI+SS-31 groups on postoperative day 3. The data are shown as the mean ± SEM. n = 5. *P < 0.05, **P < 0.01. ns indicates no significance. Figure S7. Pla2g4a overexpression doesn’t affect pyroptosis, autophagy or LMP in non-SCI mice. (A-B) Representative immunofluorescence images of p-cPLA2 and NeuN in the anterior horn areas of spinal cord from Sham, Sham+AAV-Blank, Sham+AAV-Pla2g4a groups after 3 dpi. (scale bar= 25μm). The quantification data were shown on the right. (C) cPLA2 and p-cPLA2 protein levels in the spinal cord from three groups on postoperative day 3. (D-F) Representative double immunofluorescence images for NLRP3/NeuN and Caspase-1/NeuN in spinal cord from three groups at 3 dpi (scale bar= 25μm). The quantitative integrated densities of NLRP3 and Caspase-1 in each neuron are shown on the graph. (G) Typical images of WB analyses of Caspase-1, GSDMD-N, NLRP3, NLRP1, ASC in the spinal cord. GAPDH was utilized as a loading control. (H-J) Representative double immunofluorescence images for p62/NeuN and LC3/NeuN in spinal cord lesions from three groups at 3 dpi (scale bar= 25μm). The number of LC3 II puncta and the quantitative integrated density of p62 in each neuron are shown on the graph. (K) Typical images of WB analyses of p62 and LC3 in the spinal cord. GAPDH was utilized as a loading control. (L) Comparison of the ratio of diffuse CTSL cells in the anterior horn of spinal cord among three groups. (M) Immunofluorescence staining of NeuN and CTSL in the anterior horn of spinal cord from three groups on postoperative day 3 (scale bar= 25μm). (N) Quantification results of the related protein levels from (C), (G) and (K) in the indicated groups. The data are shown as the mean ± SEM. n = 5. *P < 0.05, **P < 0.01. ns indicates no significance. Figure S8. SS-31 facilitates functional recovery in SCI mice by inhibiting cPLA2 activity. (A, B, D) Photographs of spinal cord sections in the respective groups stained with MAP2 (green) and Syn (green)/NeuN (red). Relevant quantitative results for MAP2 optical density and motor neuron-contacting synapse amounts on day 28 after SCI. (C) Spinal cord longitudinal sections from the groups at 28 days after SCI were examined via HE dyeing and Masson dyeing (scale bar = 1000 μm). (E) Quantitative analyses of Masson-positive lesions within the spinal cord in the corresponding groups. (F) Photographs of mice footprints on day 28 following spinal cord injury. Forepaw print in blue; hindpaw print in red. (G, H) Stride length (cm) and toe dragging (%) analyses of mice at 28 dpi after SCI. (I-J) Inclined plane test and BMS for the corresponding groups at day 0, 1, 3, 7, 14, 21, and 28. The data are shown as the mean ± SEM. n = 5. *P < 0.05, **P < 0.01. ns indicates no significance. Figure S9. SS-31 protected lysosome function by inhibiting p38 signal pathway. (A-D) ELISA results of the activity of lysosomal enzymes CTSD and NAGLU in lysosomal (A-B) and cytosolic (C-D) fractions extracted from the spinal cord of SCI, SCI+SS-31, and SCI+SS-31+AA groups on postoperative day 3. The data are shown as the mean ± SEM. n = 5. *P < 0.05, **P < 0.01. ns indicates no significance.

Background/PurposeMetabolic associated fatty liver disease (MAFLD) was proposed as a new definition to put emphasis on the metabolic aspects of nonalcoholic fatty liver disease (NAFLD). We aim to compare the cardiovascular and renal burden between MAFLD and NAFLD patients.Methods12183 participants were enrolled in East China. The cardiovascular burden (Framingham risk score and previous cardiovascular diseases (CVD)) and renal burden (eGFR and chronic kidney disease (CKD)) were measured.ResultsThe risk of hypertension, dyslipidemia, diabetes, overweight/obesity, and central obesity of MAFLD patients were higher than those of NAFLD. Patients with MAFLD have a similar or higher beta coefficients in Framingham risk score [beta (95%CI): male 0.062 (0.055,0.069) vs 0.041 (0.033,0.048); female 0.014 (0.012,0.016) vs 0.012 (0.01,0.014)], and higher odds ratio in previous CVD [odds ratio (95%CI): male 1.50 (1.22,1.85) vs 1.35 (1.1,1.66); female 1.58 (1.33,1.87) vs 1.45 (1.22,1.72)], compared with those with NAFLD. However, compared with males with MAFLD, the odds ratio of CKD was higher in those with NAFLD [eGFR: -2.731 (-3.422, -2.041) vs-3.578 (-4.268, -2.887). CKD: 1.44 (1.05,1.96) vs 1.56 (1.14,2.12)]. In female, CKD was only marginally associated with NAFLD [0.8 (0.62,1.02), P=0.075], but not MAFLD [0.87 (0.68,1.11), P=0.268].ConclusionsPatients with MAFLD have a similar or higher risk of future and previous CVD compared with those with NAFLD, but the risk of CKD was higher in male with NAFLD.

In this work, we revisit the Transformer-based pre-trained language models and identify two different types of information confusion in position encoding and model representations, respectively. Firstly, we show that in the relative position encoding, the joint modeling about relative distances and directions brings confusion between two heterogeneous information. It may make the model unable to capture the associative semantics of the same distance and the opposite directions, which in turn affects the performance of downstream tasks. Secondly, we notice the BERT with Mask Language Modeling (MLM) pre-training objective outputs similar token representations (last hidden states of different tokens) and head representations (attention weights of different heads), which may make the diversity of information expressed by different tokens and heads limited. Motivated by the above investigation, we propose two novel techniques to improve pre-trained language models: Decoupled Directional Relative Position (DDRP) encoding and MTH pre-training objective. DDRP decouples the relative distance features and the directional features in classical relative position encoding. MTH applies two novel auxiliary regularizers besides MLM to enlarge the dissimilarities between (a) last hidden states of different tokens, and (b) attention weights of different heads. These designs allow the model to capture different categories of information more clearly, as a way to alleviate information confusion in representation learning for better optimization. Extensive experiments and ablation studies on GLUE benchmark demonstrate the effectiveness of our proposed methods., Comment: EACL 2023(Findings)

TEG device output characteristics results

A method which can be used for standardizing the manufacturing process to fabricate an integrated temperature and relative humidity (RH) sensor based on fiber Bragg grating (FBG)and polyimide (PI) capillary is proposed. By this method, the temperature and RH cross-sensitivity problems and the influence on the humidity sensor due to the inhomogeneity of the coated moisture sensitive material film can be avoided. In this paper, two integrated temperature and RH sensors based on dual FBGs and PI capillaries with 0.03 mm and 0.05 mm wall thickness are fabricated respectively, and a measuring system is built to study the characteristics of the proposed sensors. The result shows that the temperature linearity of this sensor is above 99%, and the temperature sensitivity is higher than that of the bare FBG during temperature measurement. For the RH measurement, the sensor with 0.03 mm wall thickness PI capillary has better linearity and reproducibility, and the sensor with 0.05 mm wall thickness PI capillary has greater sensitivity and hysteresis.

For the mooring line embedded in seabed, the axial soil resistance directly affects the tension transfer characteristic along the mooring line. Effective width parameter Et was commonly adopted to assess the axial resistance of anchor chain. However, neither the chain axial resistance nor its effective width parameter in sand has been fully understood. In this two-part paper, a series of axial loading tests between chain and sand were conducted. Part I shows the monotonic loading results, and Part II aims to study the cyclic loading behavior. In Part I, a kind of novel chain-soil interaction apparatus was first developed, and the method for acquiring the axial resistance was proposed. On this basis, the influences of confining stress, chain nominal diameter, sand particle size and relative density were investigated in detail. It is found that when the chain axial resistance is fully mobilized, the effective width parameter in sand ranges from 8.91 to 11.94. The statistical average value Et=10.34 is quite close to the normalized chain perimeter Pc=10.37(3.3π). Then, an example was given to show how to calculate the chain profiles in sand using the proposed Et expression. A method for calculating the chain profile in sands was verified by the centrifuge test results. The calculation results show that higher Et values lead to larger force inclined angles at the padeye. Under the condition, the anchor will suffer larger vertical loads due to the increase of the inclination angle.

This paper addressed the problem to detect and track road boundaries of unstructured roads from images without road prior-knowledge. The unstructured roads pose the characteristic that their borders are not well defined but can be recognized either by color contrast like in rural walkways or by texture or line marks for example in urban bike paths. To have the ability to detect the road in the majority of cases, the system combines the detection of two modules, a color-based road detector, and a texture line detector. The color-based road detector works in the HSV color space and uses RGB normalization and white balance algorithms to reduce illuminations effects. Also, it uses a Manhattan distance measurement to segment the image into road pixels and a mean-shift algorithm to keep track of the road color characteristics. The texture line detector applies an inverse perspective mapping (IPM) to get rid of the perspective effects and detect parallel road borders based on a canny edge detector combined with a Hough transform. The robustness of the system is enhanced with the development of an Unscented Kalman Filter (UKF). The UKF estimates the best road borders from the measurements, also estimated borders in occlusion or in miss-detection and present a smooth road transition frame to frame. The system presents a recognition rate of 94% over 850 images within a variety of environments under various lighting conditions.

Simulation plays an important role in the development of robots, especially in the overall design stage. This paper presents a robotic simulation system, which combined USARSim (Unified System for Automation and Robot Simulation) and RCS (Real-time Control System) library innovatively. With USARSim, we can develop 3D robot platforms and environments in which the robots operate. And with RCS library, we can design a modular onboard control system simulator conveniently, which supports real time running. The framework of the simulation system is introduced, and the implementation of every components are specified in details. The application of the simulation system is demonstrated with an example. And a conclusion is given at last.

CO catalytic oxidation over Pd–Cu/Al2O3 catalysts calcined at different temperatures was studied. It was found that Pd–Cu/Al2O3-573 catalyst showed the best performance and stability than those of Pd–Cu/Al2O3-473 and Pd–Cu/Al2O3-673 catalysts. The presence of Cu2Cl(OH)3 played vital roles in achieving the highest performance over the Pd–Cu/Al2O3-573 catalyst. The preserved surface carbonate species over the Pd–Cu/Al2O3-473 catalyst and the absence of the Cu2Cl(OH)3 species over the Pd–Cu/Al2O3-673 catalyst accounted for their slight lower performances. High moisture and CO2 played negative effects on CO conversion. The former was coming from the competitive adsorption of gaseous water that hindered CO activation and benefited to the formation of carbonates under the saturated moisture conditions, and the latter might be originated from the competitive adsorption and the covering of carbonates on the catalyst surface.

The CO catalytic oxidation at ambient temperature and high space velocity was studied over the Pd\bond Cu/MOx (MOx=TiO2 and Al2O3) catalysts. The higher Brunauer–Emmett–Teller area surface of the Al2O3 support facilitates the dispersion of Pd2+ species, and the presence of Cu2Cl(OH)3 accelerates the re-oxidation of Pd0 to Pd2+ over the Pd\bond Cu/Al2O3 catalyst, which contributed to better performance of CO catalytic oxidation. The poorer activity of the Pd\bond Cu/TiO2 catalyst was attributed to the lower dispersion of Pd2+ species because of the less surface area and the non-formation of Cu2Cl(OH)3 species. The presence of saturated moisture showed a negative effect on CO conversion over the two catalysts. This might be because of the competitive adsorption, the formation of carbonate species and the transformation of Cu2Cl(OH)3 to inactive CuCl over the Pd‒Cu/Al2O3 catalyst, which facilitates the aggregation of PdO species over the Pd‒Cu/TiO2 catalyst under the moisture condition.

We propose a strain fiber sensor that is interrogated via longitudinal mode beat frequency with a linear chirped fiber Bragg grating (CFBG). Unlike conventional sensors, in the proposed sensor, the CFBG is used to convert wavelength changes into resonant cavity length changes. Further, the beat frequency stability of the fiber laser is fully utilized to significantly improve measurement accuracy. Consequently, because cavity changes account for 5.1% of the total cavity length, the measurement accuracy can be as large as 7.06×10−7 ϵ—i.e., approximately five times greater than the accuracy obtained by stretching the resonant cavity fiber. The maximum experimental strain error range is ±14.5 μϵ.

Background and Purpose Conventional subacute and chronic stroke rehabilitation programs do not provide adequate exercise to maintain the diminishing cardiovascular capacity levels. This randomised controlled trial aims to investigate the feasibility of supervised graded elastic strengthening training for patients with subacute stroke. To explore the effectiveness of strengthening and balance training that can improve the cardiovascular capacity and functional mobility in their chronic phase. Methods and Design Twenty five subjects (age: 51.3±10.8 years) with stroke in stable subacute phase were recruited at baseline and randomly allocated to receive the supervised graded elastic strengthening training (n=13) and conventional rehabilitation programs (n=12). All participants were examined by cardiac exercise testing (Marquette Case 8000 Exercise Testing System, GE, USA) to ascertain the appropriate dose of training program at baseline, 4 weeks and 12 weeks. The inpatient subjects of two groups performed the training program in hospital under the full supervision of physical therapist and their follow-up training program was supervised fully by home carer. The experimental group conducted the graded elastic strengthening training, plus conventional rehabilitation program three times per week over 12 weeks, while the control group only undertook conventional rehabilitation program. After 4 weeks training, the balance training program using Thera Band balance equipments were recommended in experimental group. The primary outcomes of this study were utilised (1) the cardiac exercise testing; (2) Fugl-Meyer; (3) 6 min walking test; (4) Berg balance scale; (5) WHOQOL-BREF; (6) muscle strength testing. Descriptive statistics were used to summarise personal demographic, stroke and baseline characteristics. Multivariate analysis of variance (MANOVA) was computed using baseline to 4 and 12 weeks change of all primary outcomes. One way ANOVA was performed to compare the difference for all outcomes measures between two groups. Results Supervised graded elastic strengthening training was feasible and safe for patients with subacute stroke. The cardiovascular capacity has significantly improved in the graded elastic strengthening training and conventional rehabilitation programs after 4 weeks and 12 weeks training (p 0.05). In multivariate analysis of variance testing the overall effect, the graded elastic strengthening training improved than conventional rehabilitation programs. Both groups increased significantly in Fugl-Meyer score, Berg balance scale and WHOQOL-BREF (p 0.05). There were significant changes for the muscle strength and 6 min walking test after 4 weeks and 12 weeks intervention in both groups (p Conclusions The present study suggested that there is preliminary evidence to determine feasibility and optimal dose of supervised graded elastic strengthening training for patients with subacute stroke. Cardiovascular capacity and functional mobility were improved using supervised elastic band and balance equipments in patients with subacute and chronic stroke.