This article implements the state‐of‐the‐art deep learning technologies for a civil engineering application, namely recognition of structural damage from images. Inspired by ImageNet Challenge and ...the development of computer hardware, the concept of Structural ImageNet is proposed herein with four naïve baseline recognition tasks: component type identification, spalling condition check, damage level evaluation, and damage type determination. A relatively small number of images (2,000) are selected from the Structural ImageNet and manually labeled according to the four recognition tasks. In order to avoid overfitting, Transfer Learning (TL) based on VGGNet (Visual Geometry Group) is introduced and applied using two different strategies, namely feature extractor and fine‐tuning. Two experiments are designed based on properties of these two strategies to find the relative optimal model parameters and scope of application. Models obtained by both strategies indicate the promising recognition results and different application potentials where feature extractor and fine‐tuning can be respectively used for preliminary analysis and for further improvement. These results also reveal the potential uses of deep TL in image‐based structural damage recognition.
Employing Deep Learning (DL) technologies to solve Civil Engineering problems is an emerging topic in recent years. However, due to the lack of labeled data, it is difficult to obtain accurate ...results with DL. One commonly used method to tackle this issue is to use affine transformation to augment the data set, but it can only generate new images that are highly correlated with the original ones. Moreover, unlike normal natural objects, distribution of structural images is much more complex and mixed. To address these challenges, Generative Adversarial Network (GAN) can be one feasible choice. We introduce one specific generative model, namely, Deep Convolutional Generative Adversarial Network (DCGAN) and propose a Leaf‐Bootstrapping (LB) method to improve the performance of this DCGAN. To effectively and quantitatively evaluate the quality of the synthetic images generated by DCGAN to complement human evaluation, Self‐Inception Score (SIS) and Generalization Ability (GA) are proposed. We also propose a pipeline based on Transfer Learning (TL) using synthetic images to help enhance a weak classifier performance under the condition of low‐data regime and limited computational resources. Finally, we conduct computer experiments with the proposed methods for two scenarios (scene level identification and damage state check) and one special synthetic data aggregation case. The results demonstrate the effectiveness and robustness of the proposed methods.
Lamb waves have multimodal and dispersion effects, which reduces their performance in damage localization with respect to resolution. To detect damage with fewest sensors and high resolution, a ...method, using only two piezoelectric transducers and based on orthogonal matching pursuit (OMP) decomposition, was proposed. First, an OMP-based decomposition and dispersion removal algorithm is introduced, which is capable of separating wave packets of different propagation paths and removing the dispersion part successively. Then, two simulation signals, with nonoverlapped and overlapped wave packets, are employed to verify the proposed method. Thereafter, with the proposed algorithm, the wave packets reflected from the defect and edge are all separated. Finally, a sparse sensor array with only two transducers succeeds in localizing the defect. The experimental results show that the OMP-based algorithm is beneficial for resolution improvement and transducer usage reduction.
Hirudin, an acidic polypeptide secreted by the salivary glands of
(also known as "Shuizhi" in traditional Chinese medicine), is the strongest natural specific inhibitor of thrombin found so far. ...Hirudin has been demonstrated to possess potent anti-thrombotic effect in previous studies. Recently, increasing researches have focused on the anti-thrombotic activity of the derivatives of hirudin, mainly because these derivatives have stronger antithrombotic activity and lower bleeding risk. Additionally, various bioactivities of hirudin have been reported as well, including wound repair effect, anti-fibrosis effect, effect on diabetic complications, anti-tumor effect, anti-hyperuricemia effect, effect on cerebral hemorrhage, and others. Therefore, by collecting and summarizing publications from the recent two decades, the pharmacological activities, pharmacokinetics, novel preparations and derivatives, as well as toxicity of hirudin were systematically reviewed in this paper. In addition, the clinical application, the underlying mechanisms of pharmacological effects, the dose-effect relationship, and the development potential in new drug research of hirudin were discussed on the purpose of providing new ideas for application of hirudin in treating related diseases.
The acoustic emission (AE) method is a popular and well-developed method for passive structural health monitoring of metallic and composite structures. The current study focuses on the analysis of ...one of its processes, sound source or signal propagation. This paper discusses the principle of plate wave signal sensing using piezoelectric transducers, and derives an analytical expression for the response of piezoelectric transducers under the action of stress waves, to obtain an overall mathematical model of the acoustic emission signal from generation to reception. The acoustic emission caused by fatigue crack extension is simulated by a finite element method, and the actual acoustic emission signal is simulated by a pencil lead break experiment. The results predicted by the mathematical model are compared with the experimental results and the simulation results, respectively, and show good agreement. In addition, the presence of obvious S0 mode Lamb waves is observed in the simulation results and experimental results, which further verifies the correctness of the analytical model prediction.
BCMA-targeting chimeric antigen receptor (CAR) T cell therapy demonstrates impressive clinical response in multiple myeloma (MM). However, some patients with BCMA-deficient tumours cannot benefit ...from this therapy, and others can experience BCMA antigen loss leading to relapse, thus necessitating the identification of additional CAR-T targets. Here, we show that FcRH5 is expressed on multiple myeloma cells and can be targeted with CAR-T cells. FcRH5 CAR-T cells elicited antigen-specific activation, cytokine secretion and cytotoxicity against MM cells. Moreover, FcRH5 CAR-T cells exhibited robust tumoricidal efficacy in murine xenograft models, including one deficient in BCMA expression. We also show that different forms of soluble FcRH5 can interfere with the efficacy of FcRH5 CAR-T cells. Lastly, FcRH5/BCMA-bispecific CAR-T cells efficiently recognized MM cells expressing FcRH5 and/or BCMA and displayed improved efficacy, compared with mono-specific CAR-T cells in vivo. These findings suggest that targeting FcRH5 with CAR-T cells may represent a promising therapeutic avenue for MM.
A fundamental feature of both early nervous system development and axon regeneration is the guidance of axonal projections to their targets in order to assemble neural circuits that control behavior. ...In the navigation process where the nerves grow toward their targets, the growth cones, which locate at the tips of axons, sense the environment surrounding them, including varies of attractive or repulsive molecular cues, then make directional decisions to adjust their navigation journey. The turning ability of a growth cone largely depends on its highly dynamic skeleton, where actin filaments and microtubules play a very important role in its motility. In this review, we summarize some possible mechanisms underlying growth cone motility, relevant molecular cues, and signaling pathways in axon guidance of previous studies and discuss some questions regarding directions for further studies.
Purpose Recent data suggest that increasing rates of hospitalization after prostate biopsy are mainly due to infections from fluoroquinolone-resistant bacteria. We report the initial results of a ...statewide quality improvement intervention aimed at reducing infection related hospitalizations after transrectal prostate biopsy. Materials and Methods From March 2012 through May 2014 data on patient demographics, comorbidities, prophylactic antibiotics and post-biopsy complications were prospectively entered into an electronic registry by trained abstractors in 30 practices participating in the MUSIC. During this period each practice implemented one or both of the interventions aimed at addressing fluoroquinolone resistance, namely 1) use of rectal swab culture directed antibiotics or 2) augmented antibiotic prophylaxis with a second agent in addition to standard fluoroquinolone therapy. We identified all patients with an infection related hospitalization within 30 days after biopsy and validated these events with claims data for a subset of patients. We then compared the frequency of infection related hospitalizations before (5,028 biopsies) and after (4,087 biopsies) implementation of the quality improvement intervention. Results Overall the proportion of patients with infection related hospitalizations after prostate biopsy decreased by 53% from before to after implementation of the quality improvement intervention (1.19% before vs 0.56% after, p=0.002). Among post-implementation biopsies the rates of hospitalization were similar for patients receiving culture directed (0.47%) vs augmented (0.57%) prophylaxis. At a practice level the relative change in hospitalization rates varied from a 7.4% decrease to a 3.0% increase. Fourteen practices had no post-implementation hospitalizations. Conclusions A statewide intervention aimed at addressing fluoroquinolone resistance reduced post-prostate biopsy infection related hospitalizations in Michigan by 53%.
Auto-regressive (AR) time series (TS) models are useful for structural damage detection in vibration-based structural health monitoring (SHM). However, certain limitations, e.g., non-stationarity and ...subjective feature selection, have reduced its wide-spread use. With increasing trends in machine learning (ML) technologies, automated structural damage recognition is becoming popular and attracting many researchers. In this paper, we combined TS modeling and ML classification to automatically extract damage features and overcome the limitation of non-stationarity. We propose a two-stage framework, namely auto-regressive integrated moving-average machine learning (ARIMA-ML) with modules for pre-processing, model parameter determination, feature extraction, and classification. Based on shaking table tests of a space steel frame, floor acceleration data were collected and labeled according to experimental observations and records. Subsequently, we designed three damage classification tasks for: (1) global damage detection, (2) local damage detection, and (3) local damage pattern recognition. The results from these three tasks indicated the robustness and accuracy of the proposed framework where 97%, 98%, and 80% average segment accuracy were achieved, respectively. The confusion matrix results showed the unbiased model performance even under an imbalanced-class distribution. In summary, the presented study revealed the high potential of the proposed ARIMA-ML framework in vibration-based SHM.
Convenient and precise assessment of the severity in coronavirus disease 2019 (COVID-19) contributes to the timely patient treatment and prognosis improvement. We aimed to evaluate the ability of ...CT-based radiomics nomogram in discriminating the severity of patients with COVID-19 Pneumonia.
A total of 150 patients (training cohort n = 105; test cohort n = 45) with COVID-19 confirmed by reverse transcription polymerase chain reaction (RT-PCR) test were enrolled. Two feature selection methods, Max-Relevance and Min-Redundancy (mRMR) and least absolute shrinkage and selection operator (LASSO), were used to extract features from CT images and construct model. A total of 30 radiomic features were finally retained. Rad-score was calculated by summing the selected features weighted by their coefficients. The radiomics nomogram incorporating clinical-radiological features was eventually constructed by multivariate regression analysis. Nomogram, calibration, and decision-curve analysis were all assessed.
In both cohorts, 40 patients with COVID-19 pneumonia were severe and 110 patients were non-severe. By combining the 30 radiomic features extracted from CT images, the radiomics signature showed high discrimination between severe and non-severe patients in the training set Area Under the Curve (AUC), 0.857; 95% confidence interval (CI), 0.775-0.918 and the test set (AUC, 0.867; 95% CI, 0.732-949). The final combined model that integrated age, comorbidity, CT scores, number of lesions, ground glass opacity (GGO) with consolidation, and radiomics signature, improved the AUC to 0.952 in the training cohort and 0.98 in the test cohort. The nomogram based on the combined model similarly exhibited excellent discrimination performance in both training and test cohorts.
The developed model based on a radiomics signature derived from CT images can be a reliable marker for discriminating the severity of COVID-19 pneumonia.