Advancing Autoencoder Architectures for Enhanced Anomaly Detection in Multivariate Industrial Time Series

In the context of rapid digitization in industrial environments,how effective are advanced unsupervised learning models,particularly hybrid autoencoder models,at detecting anomalies in industrial control system(ICS)datasets?This study is crucial because it addresses the challenge of identifying rare and complex anomalous patterns in the vast amounts of time series data generated by Internet of Things(IoT)devices,which can significantly improve the reliability and safety of these systems.In this paper,we propose a hybrid autoencoder model,called ConvBiLSTMAE,which combines convolutional neural network(CNN)and bidirectional long short-term memory(BiLSTM)to more effectively train complex temporal data patterns in anomaly detection.On the hardware-in-the-loopbased extended industrial control system dataset,the ConvBiLSTM-AE model demonstrated remarkable anomaly detection performance,achieving F1 scores of 0.78 and 0.41 for the first and second datasets,respectively.The results suggest that hybrid autoencoder models are not only viable,but potentially superior alternatives for unsupervised anomaly detection in complex industrial systems,offering a promising approach to improving their reliability and safety.

Risk evaluation of splenic hilar lymph node metastasis and survival analysis of patients with advanced gastric cancer

Background:There is no consensus regarding the influence of prophylactic no.10 lymph node(LN)dissection in patients with advanced gastric cancer(AGC).We aimed to evaluate whether patients with AGC could benefit from no.10 LN dissection and to explore the clinicopathological indicators of no.10 LN metastasis.Methods:We analyzed the data of 218 patients with AGC who underwent standard D2 lymphadenectomy(SD2;n=108)or modified D2 lymphadenectomy(MD2;n=110)between January 2017 and January 2021.In addition,we examined factors influencing no.10 LN metastasis in the SD2 group.Results:Differentiation,tumor location,and no.4 positive LNs were significantly correlated with no.10 LN metastasis(P<0.05).Borrmann classification,differentiation,depth of invasion,LN metastasis(N),and tumor size were found to correlate with survival in univariate analyses.Age,sex,extent of gastrectomy,tumor location,and extent of lymphadenectomy were not associated with survival(P>0.05).The median survival times were 72.23 and 68.56months for the SD2 andMD2 groups,respectively(P=0.635).Postoperative major morbidity and mortality rates were 37.96%and 3.70%in the SD2 group,and 23.64%and 1.82%in the MD2 group,respectively.Conclusions:Based on our findings,prophylactic no.10 lymphadenectomy may be recommended in patients with AGC who exhibit positive no.4 LN status,poor differentiation,and tumors located on the greater curvature.

Impact of time from diagnosis to chemotherapy in advanced gastric cancer: A Propensity Score Matching Study to Balance Prognostic Factors

BACKGROUND It is unclear whether treatment delay affects the clinical outcomes of chemotherapy in advanced gastric cancer(A-GC).AIM To assess whether treatment delay affects the clinical outcomes of chemotherapy in A-GC.METHODS This single-center retrospective study examined consecutive patients with A-GC between April 2012 and July 2018. In total, 110 patients with stage Ⅳ A-GC who underwent chemotherapy were enrolled. We defined the wait time(WT) as the interval between diagnosis and chemotherapy initiation. We evaluated the influence of WT on overall survival(OS).RESULTS The mean OS was 303 d. The median WT was 17 d. We divided the patients into early and elective WT groups, with a 2-wk cutoff point. There were 46 and 64 patients in the early and elective WT groups, respectively. Compared with the elective WT group, the early WT group had significantly lower albumin(Alb)levels and higher neutrophil/lymphocyte ratios and C-reactive protein(CRP)levels but not a lower performance status. The elective WT group underwent more combination chemotherapy than did the early WT group. OS was different between the two groups(230 d vs 340 d, respectively). Multivariate analysis revealed that higher CRP levels, lower Alb levels and monotherapy were significantly related to a poor prognosis. To minimize potential selection bias,patients in the elective WT group were 1:1 propensity score matched with patients in the early WT group; no significant difference in OS was found(303 d vs 311 d, respectively, log-rank P = 0.9832).CONCLUSION A longer WT in patients with A-GC does not appear to be associated with a worse prognosis.

Test Stability of Uster Advanced Fiber Information System(AFIS)

To study test stability of Advanced Fiber Information System(AFIS),card sliver produced in two experiments(12 plans in each experiment)were tested by AFIS.By a statistic analysis of the test results,the number of test times that can get a reliable test reliability(hereinafter this number of test times is referred to as Reliable Test Times,RTT)of test parameters and the coefficient of variation(CV%)values of 30 test results of each experiment plan were obtained.It's concluded that some parameters,such as length,seed coat nep(SCN)size,nep size and immature fiber content(IFC),etc.are very reliable by a test of ten or more times,but other parameters,such as SCN content,trash content,and visible foreign matter(VFM)content,etc.are not reliable until they are tested over 100 times.

Clinical significance of MET gene amplification in metastatic or locally advanced gastric cancer treated with first-line fluoropyrimidine and platinum combination chemotherapy

Objective: To investigate the clinical significance of MET gene amplification in patients with gastric cancer in the palliative setting.Methods: MET amplification was assessed using fluorescence in situ hybridization(FISH) in 50 patients and quantitative polymerase chain reaction(qPCR) in 326 patients;259 patients treated with first-line fluoropyrimidine and platinum were included for survival analysis.Results: The results of FISH and qPCR indicated that the c-MET/CEP7 ratio was correlated with gene copy number. The optimal cutoff value for the copy number using qPCR to detect MET gene amplification with FISH was 5(κ=0.778, P<0.001). Twenty-one out of 326 patients(6.4%) were identified as MET amplification with a copy number of >5 detected by qPCR. MET-amplified gastric cancer was associated with an Eastern Cooperative Oncology Group(ECOG) performance status(PS) score of ≥2(33.3% vs. 10.5% P=0.007), peritoneal metastasis(76.2% vs. 46.2%, P=0.008), and elevated bilirubin levels(28.6% vs. 7.3%, P=0.006). The median overall survival(OS) and progression-free survival(PFS) were 11.9 and 5.6 months, respectively. MET-amplified gastric cancer was not associated with survival outcomes [hazard ratio(HR)=0.68, 95% confidence interval(95% CI): 0.35-1.32,P=0.254 for PFS;HR=0.68, 95% CI: 0.35-1.32, P=0.251 for OS].Conclusions: qPCR can be used to detect MET gene amplification. MET amplification was not a predictor of poor prognosis in patients with metastatic or unresectable gastric cancer.

Professor YANG Yu-fei’s Experience on the Integrated Chinese and Western Medicine Treatment Strategy for Advanced Colorectal Cancer

Colorectal cancer is one of the common malignant tumors in China.It poses a serious threat to the national health of China.For advanced colorectal cancer, the main goal of treatment is to prolong survival and improve quality of life.It complements other advantages, showing good therapeutic results.However, how to grasp the timing of integrated Chinese and Western Medicine for the treatment of advanced colorectal cancer and use the integrated Chinese and Western Medicine treatment methods flexibly contains profound therapeutic art.Prof.YANG Yu-fei is an authoritative expert in the field of integrated Chinese and Western medicine for colorectal cancer.She is good at accurately grasping the timing of treatment of integrated Chinese and Western Medicine, and flexibly adjusts the treatment strategy according to the specific conditions.In this paper, we shared Professor YANG Yu-fei's strategy for treating advanced colorectal cancer with emphasis on integrated Chinese and Western Medicine and attached a typical case, with a view to provide reference for the treatment of advanced colorectal cancer with integrated Chinese and Western Medicine.

Applications of advanced signal processing and machine learning in the neonatal hypoxic-ischemic electroencephalography

Perinatal hypoxic-ischemic-encephalopathy significantly contributes to neonatal death and life-long disability such as cerebral palsy. Advances in signal processing and machine learning have provided the research community with an opportunity to develop automated real-time identification techniques to detect the signs of hypoxic-ischemic-encephalopathy in larger electroencephalography/amplitude-integrated electroencephalography data sets more easily. This review details the recent achievements, performed by a number of prominent research groups across the world, in the automatic identification and classification of hypoxic-ischemic epileptiform neonatal seizures using advanced signal processing and machine learning techniques. This review also addresses the clinical challenges that current automated techniques face in order to be fully utilized by clinicians, and highlights the importance of upgrading the current clinical bedside sampling frequencies to higher sampling rates in order to provide better hypoxic-ischemic biomarker detection frameworks. Additionally, the article highlights that current clinical automated epileptiform detection strategies for human neonates have been only concerned with seizure detection after the therapeutic latent phase of injury. Whereas recent animal studies have demonstrated that the latent phase of opportunity is critically important for early diagnosis of hypoxic-ischemic-encephalopathy electroencephalography biomarkers and although difficult, detection strategies could utilize biomarkers in the latent phase to also predict the onset of future seizures.

Parameters Identification of Continuous-Time Hammerstein System with Advanced Gauss Pseudo spectral Method

An advanced Gauss pseudospectral method(AGPM) was proposed to estimate the parameters of the continuous-time(CT)Hammerstein model.The nonlinear part of the Hammerstein system is approximated with pseudospectral approximation method.The linear part was written as a controllable canonical form to circumvent the high order time-derivative of the input and output(I/O) signals,which could multiply the measurement noise in the identification procession.Furthermore,an output error minimization was constructed for the CT Hammerstein model identification,which was then transcribed into a nonlinear programming(NLP) problem by AGPM.AGPM could converge to the true values of the CT Hammerstein model with few interpolated Legendre-Gauss(LG) nodes.Lastly,two illustrative examples were proposed to verify the accuracy and efficiency of the method.

电网电压跌落下基于二分迭代和NITA法的同步调相机全域瞬态流热分析

同步调相机安装在特高压输电系统换流站内用于稳定输电系统的电压,输电系统可能发生电压跌落故障决定其需要具备工作在极限运行状态的能力。温升是制约调相机极限调节能力的重要因素之一,因此,对同步调相机极限运行时的瞬态流热分析非常必要。该文建立调相机与输电系统耦合的暂态分析模型,研究电网电压跌落时调相机的瞬态响应特性。为综合评价电网电压跌落故障对调相机热表征的影响,建立300 Mvar同步调相机全域内流固耦合瞬态分析模型;提出一种二分迭代的流固耦合模型边界确定方法,通过数以亿计的多面体网格离散(共1.4亿网格单元),同时,基于非迭代时间推进法对全域复杂传热模型进行求解,研究调相机多风腔内流体和温度相互影响机理;探讨极限运行状态下温度约束对调相机安全运行的影响。为对该文所提研究方法进行验证,开展调相机总装温升实验,证明其可行性与准确性。

Application of advanced oxidation processes for removing salicylic acid from synthetic wastewaters

In this study,advanced oxidation processes(AOPs) such as anodic oxidation(AO),UV/H_2O_2 and Fenton processes(FP) were investigated for the degradation of salicylic acid(SA) in lab-scale experiments.Boron-doped diamond(BDD) film electrodes using Ta as substrates were employed for AO of SA.In the case of FP and UV/H_2O_2,most favorable experimental conditions were determined for each process and these were used for comparing with AO process.The study showed that the FP was the most effective process under aci...

Clinical management of advanced gastric cancer: The role of new molecular drugs

Gastric cancer is the fourth most common malignant neoplasm and the second leading cause of death for cancer in Western countries with more than 20000 new cases yearly diagnosed in the United States. Surgery represents the main approach for this disease but, notwithstanding the advances in surgical techniques, we observed a minimal improvement in terms of overall survival with a significant increasing of relapsing disease rates. Despite the development of new drugs has significantly improved the effectiveness of chemotherapy, the prognosis of patients with unresectable or metastatic gastric adenocarcinoma remains poor. Recently, several molecular target agents have been investigated; in particular, trastuzumab represents the first target molecule showing improvements in overall survival in human epithelial growth factor 2-positive gastric cancer patients. New molecules targeting vascular epithelial growth factor, mammalian target of rapamycin, and anti hepatocyte growth factor-c-Met pathway are also under investigation, with interesting results. Anyway, it seems necessary to select more accurately the population to treat with new agents by the identification of new biomarkers in order to optimize the results. In this paper we review the actual &#x0201c;scenario&#x0201d; of targeted treatments, also focusing on the new agents in development for gastric cancer and gastro-esophageal carcinoma, discussing their efficacy and potential applications in clinical practice.

Impact of interval between neoadjuvant chemoradiotherapy and surgery in rectal cancer patients

BACKGROUND Epidemiologically,in China,locally advanced rectal cancer is a more common form of rectal cancer.Preoperative neoadjuvant concurrent chemoradiotherapy can effectively reduce the size of locally invasive tumors and improve disease-free survival(DFS)and pathologic response after surgery.At present,this modality has become the standard protocol for the treatment of locally advanced rectal cancer in many centers,but the optimal time for surgery after neoadjuvant therapy is still controversial.AIM To investigate the impact of time interval between neoadjuvant therapy and surgery on DFS and pathologic response in patients with locally advanced rectal cancer.METHODS A total of 231 patients who were classified as having clinical stage II or III advanced rectal cancer and underwent neoadjuvant chemoradiation followed by surgery at the National Cancer Center/Cancer Hospital,Chinese Academy of Medical Sciences and Peking Union Medical College from November 2014 to August 2017 were involved in this retrospective cohort study.The patients were divided into two groups based on the different time intervals between neoadjuvant therapy and surgery:139(60.2%)patients were in group A(≤9 wk),and 92(39.2%)patients were in group B(>9 wk).DFS and pathologic response were analyzed as the primary endpoints.The secondary endpoints were postoperative complications and sphincter preservation.RESULTS For the 231 patients included,surgery was performed at≤9 wk in 139(60.2%)patients and at>9 wk in 92(39.8%).The patients’clinical characteristics,surgical results,and tumor outcomes were analyzed through univariate analysis combined with multivariate regression analysis.The overall pathologic complete response(pCR)rate was 27.2%(n=25)in the longer time interval group(>9 wk)and 10.8%(n=15)in the shorter time interval group(≤9 wk,P=0.001).The postoperative complications did not differ between the groups(group A,5%vs group B,5.4%;P=0.894).Surgical procedures for sphincter preservation were performed in 113(48.9%)patients,which were not significantly different between the groups(group A,52.5%vs group B,43.5%;P=0.179).The pCR rate was an independent factor affected by time interval(P=0.009;odds ratio[OR]=2.668;95%CI:1.276-5.578).Kaplan-Meier analysis and Cox regression analysis showed that the longer time interval(>9 wk)was a significant independent prognostic factor for DFS(P=0.032;OR=2.295;95%CI:1.074-4.905),but the time interval was not an independent prognostic factor for overall survival(P>0.05).CONCLUSION A longer time interval to surgery after neoadjuvant therapy may improve the pCR rate and DFS but has little impact on postoperative complications and sphincter preservation.

Study on Advanced Water Injection Time in Low Permeability Reservoir

A certain formation pressure level must be kept due to the threshold pressure of the low-permeability reser-voir during the seepage. Advanced water injection can keep the formation pressure at a higher level and keep a higher pressure gradient, which is an effective way to develop low-permeability reservoir. Based on the mechanism of advanced water injection and characteristic of porous flow in low permeability reservoir, a seepage model considering threshold pressure gradient is established to determine the formation pressure distribution at anytime as the water is injected at a constant speed. The optimum water injection time for the advanced water injection technology can be determined by using this model. The calculated result coincides basically with the numerical simulation result, which indicates that the model put forward in this paper is feasible.

Forced Oscillation of Neutral Advanced Dynamic Equations with Positive and Negative Coeffcients on Time Scales

In this paper,we consider the forced neutral advanced dynamic equations with positive and negative coeffcients[x(t)-R(t)x(t+r)]+P(t)x(t+τ)△Q(t)x(t+δ)= f(t),t≥t0 on a time scale T.Several suffcient conditions are obtained for the oscillation of the dynamic equations.The main result is illustrated by an example.

Timing of advanced water flooding in low permeability reservoirs

It is very important to design the optimum starting time of water injection for the development of low permeability reservoirs. In this type of reservoir the starting time of water injection will be affected by a reservoir pressure-sensitive effect. In order to optimize the starting time of water injection in low permeability reservoirs, this effect of pressure change on rock permeability of low permeability reservoirs was, at first, studied by physical simulation. It was shown that the rock permeability decreases exponentially with an increase in formation pressure. Secondly, we conducted a reservoir engineering study, from which we obtained analytic relationships between formation pressure, oil production rate, water production rate and water injection rate. After our physical, theoretical and economical analyses, we proposed an approach which takes the pressure-sensitive effect into consideration and designed the optimum starting time of water injection, based on the principle of material balance. Finally, the corresponding software was developed and applied to one block of the Jiangsu Oilfield. It is shown that water injection, in advance of production, can decrease the adverse impact of the pressure-sensitive effect on low permeability reservoir development. A water-flooding project should be preferably initiated in advance of production for no more than one year and the optimum ratio of formation pressure to initial formation pressure should be maintained at a level between 1.05 and 1.2.

Timed automata for metric interval temporal logic formulae in prototype verification system

Based on analysis of the syntax structure and semantics model of the metric interval temporal logic （MITL） formulas, it is shown how to transform a formula written in the real-time temporal logic MITL formula into a fair timed automaton （TA） that recognizes its satisfying models with prototype verification system （PVS） in this paper. Both the tabular construction＇s principles and the PVS implementation details are given for the different type of MITL formula according to the corresponding semantics interpretations. After this transformation procedure, specifications expressed with MITL formula can be verified formally in the timed automata framework developed previously.

基于Runge-Kutta的自回归物理信息神经网络求解偏微分方程

物理信息神经网络离散时间模型(PINN-RK)是深度学习技术与龙格库塔方法相结合的产物,在求解偏微分方程时具有非常出色的稳定性和较高的求解精度.但是,受到龙格库塔算法本身的限制,PINN-RK模型仅能实现单步时间预测,且计算效率较低.因此,为了实现多时间步长预测和提高模型的计算效率,提出了一种基于龙格库塔法的自回归物理信息神经网络模型(SR-PINN-RK).该模型基于自回归时间步进机制,改进了神经网络的训练流程和网络结构,相比PINN-RK模型,大幅减少了神经网络的训练参数,提高了模型的计算效率.此外,在自回归机制的作用下,该模型通过对标签数据的动态更新,成功实现了对偏微分方程解的多时间步长预测.为了验证文中模型的求解精度和计算效率,分别求解了Allen-Cahn方程和Burgers方程,并与文献中的基准解进行了对比.结果表明,模型预测解与基准解之间具有很高的一致性,求解Allen-Cahn方程和Burgers方程的最大相对误差均低于0.009.

Survival Effect of Tube Feeding in Patients with Advanced Dementia and Comorbidities

Patients suffering from advanced dementia who have difficulties in eating and swallowing are often tube fed (TF). They suffer, on most occasions, from additional diseases. The effect of tube feeding in advanced dementia patients is controversial and is largely discussed in the medical literature. The purpose of the present study, is to report the effect of tube feeding on the survival time in patients with advanced dementia who suffered from additional diseases. One hundred and forty nine patients with Functional Assessment Stage of Alzheimer Disease (FAST) stage 5>, hospitalized in skilled geriatric nursing department between July 2008 and December 2011 were included in the study. Oral and tub feeding was decided following the evaluation by speech therapists at admission. 99 (66.5%) of the patients were tube fed and 50 (33.5%) received oral feeding at admission. The mean age was 80 ± 11. Sixty four percent of the patients were admitted from general hospitals. Comorbidities were significantly higher in the tube fed group, median survival time was 164 days. The median survival time of the OF group was 78 days (χ2 = 0.94;p = 0.33). Prevalence of tube feeding increased significantly from 53.85% to 90.9% in patients with 3 risk factors such as pressure ulcers, low Hgb and low BMI. The survival time of the tube fed patients was longer compared to patients who were orally fed. Tube feeding in patients with dementia continues to be controversial, and additional researchable data are helpful in the decision to insert a tube.

Carry out an Independent Policy and Advance with the Times

The topic of "China’s foreign policy under new circumstances" for discussion raised by Experts Forum of Contemporary International Relations is very important, because situation determines tasks, tasks produce policies and foreign policy is the continuation of domestic policy.

Advanced production decline analysis of tight gas wells with variable fracture conductivity

Considering the characteristics that the fracture conductivity formed by hydraulic fracturing varies across space and time, a new mathematical model was established for seepage flow in tight gas fractured vertical wells which takes into account the effects of dual variable conductivity and stress sensitivity. The Blasingame advanced production decline curves of the model were obtained using the finite element method with hybrid elements. On this basis, the effects of fracture space and time dual variable conductivity and stress sensitivity on Blasingame curve were analyzed. The study shows that the space variable conductivity mainly reduces decline curve value at the early stage; the time variable conductivity can result in drops of the production and the production integral curves, leading to a S-shaped curve; dual variable conductivity is the superposition of the effects given by the two variable conductivities; both time and space variable conductivities cannot delay the time with which the formation fluid flow reaches the quasi-steady state. The stress sensitivity reduces the curve value gradually rather than sharply, delaying the time the flow reaching the quasi-steady state. Ignoring the effects of variable conductivity and stress sensitivity will not affect the estimation on well controlled dynamic reserves. However, it can result in large errors in the interpretation of fractures and reservoir parameters. Conventional advanced production decline analyses of a tight gas fractured well in the Sulige gas field showed that the new model is more effective and reliable than the conventional model, and thus it can be widely applied in advanced production decline analysis of wells with the same characteristics in other gas fields.