Valve Status Research Articles

Hydraulic valve fault diagnosis based on EEWT and POA-KELM

Abstract With the wide application of hydraulic systems in the industrial and mechanical fields, the analysis and study of hydraulic valve failures have become increasingly important. This study aims to identify and analyze possible faults by analyzing the operating status of hydraulic valves under different loads. In this study, we propose using a combination of two phases for the failure analysis of hydraulic valves. The first stage is the signal processing stage, in which the pressure signals of hydraulic valves under different loads are collected, the pressure signals are decomposed using the extended empirical wavelet transform (EEWT) to obtain a series of sub-signal components, and the 17 eigenvectors of these components are computed, and the feature selection is carried out using the method of sequence backward selection. Next is the classification stage, where the features selected in the first stage are input into a kernel extreme learning machine (KELM) classifier improved by the Pelican optimization algorithm (POA) to classify and identify the anomalous signals. The experimental results show that the fault diagnosis method based on EEWT and POA-KELM can effectively detect and recognize hydraulic valve faults.

Impact of atrial fibrillation on mortality and non-fatal clinical outcomes in 28 492 patients who had undergone open-heart aortic versus mitral valve surgery: a statewide population-linkage study

Abstract Background Postoperative atrial fibrillation (AF) is common after cardiac surgery and has been associated with increased mortality and morbidity. Few studies have defined the impact of AF status (New-AF versus Prior-AF) on clinical outcomes during long-term follow-up and compared AF-related clinical outcomes between patients who had undergone isolated aortic valve surgery (AVSx) or mitral valve surgery (MVSx). Purpose This study aims to examine the impact of AF status on clinical outcomes in patients who had undergone AVSx versus MVSx using a statewide-population administrative database. Methods Patients who had open-heart cardiac valve surgery (1 January 2003 to 31 March 2021) identified from the Admitted-Patient-Data-Collection database in an Australian state, were stratified by AF status (No-AF vs New-AF vs Prior-AF during index admission for valve surgery) and followed up to 31 March 2022. A linkage look-back to year-2001 was performed to identify prior-AF history. Kaplan-Meier and multivariable Cox regression were performed to assess the impact of AF status on all-cause mortality. Fine-Gray competing risk analysis to account for the competing death events was used to assess non-fatal clinical outcomes. Results The overall cohort comprised 28 492 patients (median age 71.6yrs [interquartile range (IQR) 62.7–78.3yrs; 65.6% males): AVSx, n=18949; MVSx, n=9543 (Table 1). During a median follow-up of 6.6yrs (3.42–10.5yrs), Prior-AF and New-AF patients had significantly higher all-cause mortality (AVSx – Prior-AF: 1771 (57.9%) vs New-AF: 2854 (41.5%) vs No-AF: 2961 (32.8%); MVSx – Prior-AF: 1126 (41.0%) vs New-AF: 1034 (29.8%) vs No-AF: 747 (22.4%) (both Kaplan-Meier logrank P<0.001). After adjusting for differences in baseline characteristics and admission year-groups, in the AVSx subgroup, both new-AF and prior-AF were independently associated with all-cause mortality (aHR=1.16, 95%CI=1.10–1.22; aHR=1.69, 95%CI=1.59–1.79 respectively, both P<0.001) compared to no-AF patients. In the MVSx subgroup, only prior-AF was associated with increased risk of all-cause mortality (aHR=1.47, 95% CI=1.33–1.61, P<0.001). Competing risk analyses showed higher rehospitalisation rates for AF and congestive cardiac failure for new-AF and prior-AF groups compared to No-AF patients after both AVSx and MVSx, whereas the rehospitalisation rate for ischemic stroke was only significantly higher in the AVSx new-AF and prior-AF subgroups (all p < 0.05) (Table 2). Conclusions Patients undergoing open-heart aortic or mitral valve surgery are at risk of significant adverse clinical outcomes including death, heart failure and ischaemic stroke, with the target valve and baseline AF status having a differential impact on clinical outcomes. Drivers behind these high rates of adverse outcomes should be explored further and strategies should be developed and tailored accordingly to different surgical patient groups to improve their prognosis.

Design and Research of Intelligent Valve Control

The design and research of intelligent valve control system aims to improve the efficiency and accuracy of fluid control in industrial process automation. By integrating advanced sensing technology, communication protocols and intelligent algorithms, the precise control of valve position, pressure and flow rate is achieved. STM32 controller and a variety of sensors are used in the system to realize real-time monitoring and feedback control of valve status. The control strategy based on PID algorithm enables the valve to maintain stable operating performance under complex working conditions. In addition, the safety and reliability of the system are also considered, and the robustness of the system is improved by redundant design and fault detection mechanism. The experimental results show that the designed intelligent valve control system has significant advantages in improving response speed, control precision and energy efficiency, and has a wide range of industrial application prospects.

A two-year dataset of energy, environment, and system operations for an ultra-low energy office building

This paper describes a two-year high-fidelity dataset for an ultra-low energy office building and living laboratory called HouseZero®. The building integrates multiple low-energy technologies, such as natural ventilation with automatic windows, ground source heat pump, and thermally activated building systems. The building’s performance is continuously monitored with an extensive sensor network. The dataset consists of breakdown energy end uses, photovoltaic (PV) production, zone-level indoor environment including indoor air temperature, CO2 concentration, and relative humidity, micro-climatical conditions, building façade temperature, and detailed system operations including zone-level BTU meter, valve status, slab temperature, window/skylight opening status, heat pump, and geothermal well operations. The data can be used to support data analytics of ultra-low-energy building operations, and data-driven modeling of low-energy building systems.

Exposure and outcomes of aortic valve change in patients initiating dialysis.

Aortic stenosis (AS) and aortic valve calcification (AVC) are occasionally observed in patients receiving maintenance dialysis. However, their prevalence and factors associated with them in patients undergoing dialysis remain unknown. We aimed to elucidate the aortic valve status at the time of dialysis initiation and patient prognosis based on aortic valve status. We analyzed 289 patients initiating dialysis (hemodialysis: peritoneal dialysis = 275:14) between 2016 and 2023. "AS and/or AVC" was detected using echocardiography. AS was defined as a maximum transaortic velocity > 2.0m/s. Statistical analyses including multivariable logistic regression and Cox regression were used to assess the association between patient characteristics and survival outcomes. Aortic valve changes were observed in 121 (42%) patients, among which 33 (11%) met the AS criteria. The mean age of patients in the AS, AVC without AS, and control groups was 79.1 ± 8.9, 75.9 ± 9.2, and 68.3 ± 12.9, respectively (P < 0.001). Multivariable logistic regression models showed that only age was associated with aortic valve changes (P < 0.001). Age and other important factor-adjusted multivariable Cox regression models showed that AS was an independent risk factor for death after dialysis initiation (hazard ratio (HR): 1.95, 95% confidence interval (CI): 1.06 - 3.59, P = 0.04). However, aortic valve changes ("AS and/or AVC") were not a risk factor for death (HR: 1.51, 95% CI 0.95 - 2.39, P = 0.08). With the growing older population undergoing dialysis, aortic valve changes should be closely monitored. Particularly, AS is crucial because of its impact on patient prognosis.

A Real-Time Intelligent Valve Monitoring Approach through Cameras Based on Computer Vision Methods.

Abnormal valve positions can lead to fluctuations in the process industry, potentially triggering serious accidents. For processes that frequently require operational switching, such as green chemical processes based on renewable energy or biotechnological fermentation processes, this issue becomes even more severe. Despite this risk, many plants still rely on manual inspections to check valve status. The widespread use of cameras in large plants now makes it feasible to monitor valve positions through computer vision technology. This paper proposes a novel real-time valve monitoring approach based on computer vision to detect abnormalities in valve positions. Utilizing an improved network architecture based on YOLO V8, the method performs valve detection and feature recognition. To address the challenge of small, relatively fixed-position valves in the images, a coord attention module is introduced, embedding position information into the feature channels and enhancing the accuracy of valve rotation feature extraction. The valve position is then calculated using a rotation algorithm with the valve's center point and bounding box coordinates, triggering an alarm for valves that exceed a pre-set threshold. The accuracy and generalization ability of the proposed approach are evaluated through experiments on three different types of valves in two industrial scenarios. The results demonstrate that the method meets the accuracy and robustness standards required for real-time valve monitoring in industrial applications.

In-line valve condition assessment by transient analysis

Negligently partially closed in-line valves in pipes cause undesirable minor head losses that reduce the carrying capacity and increase operating costs attributable to energy consumption. However, few obtained results of in-line valves condition assessment has been offered in the literature. The paper makes further investigations on the interaction between the valve and the transient pressure wave, which provides experimental data support and reference opinions for the transient wave detection of in-line valve status with practical applications. The position of the in-line valve is evaluated by a simple equation and the propagation of transient pressure wave was analyzed. The relevant equations are derived and correlation analysis was performed. Based on the pressure increase value, the head loss coefficient or valve opening degree can be determined, and feasible suggestions for detecting the valve opening degree using transient waves have been made. Under the selected experimental conditions, the absolute error between the calculated valve opening degree and the actual opening degree is below 0.06 (1 for fully open valve opening and 0 for fully closed valve opening), proving that the valve opening assessment method is effective. This paper demonstrates the application of transient wave technology in real-time monitoring and assessment of valve positions. Through this technology, the valve positions in water distribution systems can be optimized effectively, thereby reducing resistance losses, meeting varying water demands, avoiding unnecessary energy waste, and enhancing the overall energy efficiency of the water supply system.

Monitoring and evaluation of the status of Motor-Operated Valves in nuclear power plants

To address the inability to monitor the performance status of Motor-Operated Valves (MOV) during the operation of nuclear power plants, develop the MOV status evaluation model and design a valve online monitoring system. During valve operation, the data acquisition unit (DAU) installed in the distribution cabinet of the motor control center (MCC) can capture the voltage, current and switch signals. On the basis of the aforementioned parameters, the active power curve and characteristic points of the valve can be determined, implemented monitoring and evaluation of the status and performance of MOV.

Abstract 16826: A Rare Case of an Emergent Third Valve in Valve Intervention

Introduction: Bioprosthetic valves are prone to structural degeneration due to an average lifetime of 10-15 years. Valve in Valve (VIV) TAVR has become an alternative to the high-risk surgical patients as the growing number of successful bioprosthetic valve placements and the deteriorations over time. Emergent VIV intervention effectiveness with prior multiple artificial valves is less studied. Hypothesis: We report a case of a 54-year-old male presenting with obstructive shock with history of substance abuse and bicuspid aortic valve status post bioprosthetic aortic valve with aortic root repair at age of 15(1983), aortic stenosis requiring repeat surgical bioprosthetic Aortic Valve Replacement (SAVR) (25 mm) at age 42(2011). Recalcitrant cardiogenic shock despite vasopressor. Emergent Heart catheterization showed mild coronary artery disease (CAD) and severe cardiomyopathy, Left Ventricle ejection Fraction 15-20%. Aortic valve Gradient was 80 mmHg. Transesophageal Echocardiography (TEE) ruled out endocarditis. Impella CP and Swan-Ganz catheter were placed for hemodynamic support and monitoring. Methods: The Heart Team approach was followed. Patient was at high risk of mortality and operative complications with redo sternotomy. He subsequently underwent valve in valve TAVR with self-expanding 26 mm valve. Prior surgical valve did not require fracturing post implantation. Results: Post Trans Aortic Valve Replacement (TAVR) the mean gradient on echocardiogram decreased to 10 mmHg and LVEF recovered to 55-60%. The patient was successfully weaned off Vaso-pressors and eventually discharged home. Conclusion: Most common cause of transcatheter bioprosthetic valve failure is stenosis from calcification and thrombosis. VIV TAVR in recent studies has proven to be safe with favorable outcomes. Our patient had already undergone two surgical valve interventions, presented with critical bioprosthetic valve stenosis requiring salvage TAVR for cardiogenic shock. After the Heart team approach, TAVR was opted as a more favorable option for the patient. Although, the life of a bioprosthetic valve results in multiple interventions but overall valve in Valve technique has been proven to provide expected results in emergent high-risk patients.

A smartphone-controllable molecular diagnostic platform for SARS-CoV-2 detection by reverse-transcription recombinase polymerase amplification

We constructed a portable molecular diagnostic microsystem, enabling the whole processes of genetic analysis for SARS-CoV-2. The proposed platform was a miniaturized size and was comprised of three main parts: a microfluidic device with a 3D-printed solution loading cartridge, an operating system including a tailor-made heater board, a syringe pump and an optical fluorescence detector, and a smartphone with a customized app for the remote and automatic control. The microfluidic device was designed to control the complicated fluidic flow through only the passive valves, and could perform the RNA extraction from the viral lysate via glass membranes and the target gene amplification by reverse-transcription recombinase polymerase amplification (RT-RPA). The 3D-printed cartridge allowed the facile chemical solution storage and loaded a designated solution into the microfluidic device. A smartphone with a browser-based app could control the on-and-off status of solenoid valves, the workflow of a syringe pump to automatically introduce solutions on a chip, the heater board to supply the constant temperature for RT-RPA, and the periodic activation of the Pi camera and a 488 nm laser to measure the fluorescence signal in the reaction chambers that showed the amplification profiles on the smartphone display in real-time. Using the proposed smartphone-controllable diagnostic platform, the N gene and S gene of SARS-CoV-2 were successfully detected at a limit-of-detection (LOD) of 10 copies/μL within 1 hr in a sample-to-answer-out format. Ten out of 22 clinical samples were successfully analyzed for SARS-CoV-2, demonstrating the practical applications for the on-site diagnostics of patient samples.

An All-Electric Gate Valve Actuator for Subsea Production Control Systems, Part A: Prototype and Test

To bridge the gap that exists in the key equipment of the new subsea production control system, the all-electric subsea gate valve actuator, and exploit subsea oil and gas resources with high reliability and safety while saving energy, this paper proposes a novel concept prototype of an all-electric subsea gate valve actuator which has the key functions of a redundant drive, failsafe closing, auxiliary override, position indication, and low-power position holding. It satisfied the electrically-driven requirements of the subsea gate valves and achieved Safety Integrity Level 3. The prototype was developed and tested successfully. The all-electric subsea gate valve actuator is suitable for controlling subsea gate valves with various sizes and rated working pressures to minimize the power consumption for the purpose of keeping the valves open and safely closing them in the event of the electrical failure. An override and position-indicating mechanism is equipped for emergency operation and the visual indication of the status of subsea gate valves.

The V-shaped double-layer patch technique for complete atrioventricular septal defect: A novel surgical technique.

Several surgical techniques for repair of a complete atrioventricular septal defect have been developed. However, the postoperative complications with these methods may lead to reoperation during follow-up. The aim of this report is to share our experience with a modified surgical technique for complete atrioventricular septal defect that has anatomic advantages postoperatively and could reduce the reoperation rate. Twenty-nine patients who underwent repair of complete atrioventricular septal defect using a V-shaped double-layer patch between April 2011 and September 2019 were retrospectively investigated. There were no deaths (0%) and only 1 reoperation (3.4%) in the series. The aortic crossclamp and cardiopulmonary bypass times were 62.7±16.0minutes and 113.9±25.9minutes, respectively. The median follow-up duration was 5.1years. To date, no significant residual ventricular septal defects have been detected and no left ventricular outflow tract obstruction has been seen on echocardiography in any patient. During follow-up, the left atrioventricular valve status was assessed as no incompetence in 9 patients (31.0%), trivial in 18 patients (62.1%), and mild in 2 patients (6.9%). The V-shaped double-layer patch technique is a valuable surgical option for patients with complete atrioventricular septal defects. The midterm results in our series document excellent performance of this technique, which augments the area of the anterior valve of the left atrioventricular valve to make it closer to a normal mitral valve and may also reduce the need for reoperation.

GEMELLA MORBILLORUM BIOPROSTHETIC AORTIC VALVE INFECTIVE ENDOCARDITIS WITH SEPTIC SHOCK CAUSING LEAFLET PERFORATION AND AORTIC ROOT ABSCESS

GEMELLA MORBILLORUM BIOPROSTHETIC AORTIC VALVE INFECTIVE ENDOCARDITIS WITH SEPTIC SHOCK CAUSING LEAFLET PERFORATION AND AORTIC ROOT ABSCESS

Correlation of Urethral Ratio and Bladder Wall Thickness with Cystoscopic Findings in Posterior Urethral Valve Patients to Assess Residual Valves.

Introduction:Posterior urethral valve (PUV) is life threatening congenital anomaly of urinary tract. Aim of the study was to correlate urethral ratio (UR) and bladder wall thickness (BWT) with cystoscopic findings in PUV patients to assess residual valves and to validate UR as a diagnostic tool for residual valves. It also aimed to assess the utility of bladder wall thickness in diagnosis of residual valves.Materials and Methods:A total of 31 patients were included in the prospective study done from 2017 to 2019. Calculation of UR was done in oblique VCUG films by dividing maximum posterior and anterior urethral diameter without the catheter insitu. Measurement of BWT was done at dome and bladder neck at full distension with feeding tube insitu and was done at same volume in follow up. The procedure was repeated at 3 months follow up. The findings were compared with cystoscopic findings for the status of residual valves as gold standard. Each patient served as control for self in the study.Results:Median age of presentation was 1 years with range of 1day to 10 years. The most common complains at the time of presentation in our study was straining (35.48%) followed by antenatally diagnosed patients (25.81%) and recurrent UTI (19.36%). Pre fulguration median UR was 2.45. Post Fulguration median UR was 1.20. It showed a statistically significant reduction (p < 0.001) after fulguration. Pre fulguration median BWT was 4 mm. Post fulguration median BWT was 2.5 mm. BWT showed a statistically significant reduction (p < 0.001) after fulguration as well. ROC curve was plotted for UR and BWT. BWT more than 1.95 mm (sensitivity-80%) and UR more than 1.2 (sensitivity-70%) indicates residual valves.Conclusion:A step ladder approach including BWT, UR and check cystoscopy can serve as a new diagnostic algorithm for the assessment of residual valves thereby avoiding extra radiation and general anesthesia exposure.

Continuous Status Monitoring of Industrial Valve Using OCC-Enabled Wireless Sensor Network

This technical article proposes an instrumentation valve (IV) status monitoring system based on optical camera communication (OCC). A transmitter circuit equipped with a temperature sensor is integrated into each IV. Each transmitter also includes two light-emitting diodes (LEDs) forming an LED group, where one of the LEDs is used to transmit the data. During the reception process, each LED group is recognized and classified using a neural network. Then, the LEDs are individually identified using the region-of-interest detection mechanism. A close circuit television (CCTV) camera is employed to receive the data, which are henceforth stored in a cloud server for further monitoring. An angle measurement algorithm capable of determining the angle generated due to unwanted failure related to the proper closing or opening of the IV is proposed. Additionally, a neural blind deconvolution algorithm is proposed to alleviate the blur effect in the received images. In short, the data transmitted by the LED contain information including valve ID, temperature, and the amount of inflection produced by the IV. The entire OCC system is implemented, and its performance is evaluated using Python 3.7. Data with blurry images are successfully received up to a communication distance of 10 m, whereas the lowest bit error rate (BER) is achieved at a communication distance of 2 m.

The Parametric Optimization of Voltage Multiplier

The evolutionary model of voltage multiplier parametric optimization which includes 5 diodes and 5 capacitors is reviewed. It executes the transformation of alternating into constant voltage using a five times larger amplitude. The valve work is modelled according to the scheme of an ideal key. The original mathematical model of voltage multiplier which includes additional logical variables is deducted. It aссepts binary meanings 0 and 1, where 0 corresponds to closed valve status and 1 corresponds to open. In order to receive such a model, it is necessary to indicate the amount of open and closed valve combinations. Then for each of them, it is necessary to write the system of differential equations. Comparing them with each other the single differential equation system with additional logical variables is written as a generalization. The evolutional model is used in order to select the capacitor volume meaning. The goal function forecasts two conditions: maximum meaning of output voltage 1 kV and its minimal fluctuations in the stable regime.

Prevalence of bicuspid aortic valve in patients with aneurysmal sub-arachnoid hemorrhage: neurologic features and outcomes

Abstract Background Intracranial aneurysm (IA) has been associated with bicuspid aortic valve (BAV) with prevalence of 6–10% which has led some to recommend routine surveillance for IA in BAV patients. Higher incidence of IA in BAV could signal towards systemic arteriopathy beyond the thoracic aorta. With CoA as an independent risk factor for IA formation, it is currently uncertain if the association between IA and BAV reflects the association of BAV and CoA, or the result of BAV-related systemic arterial changes. Purpose Despite this association, 2 small studies suggest no increased risk of the most important patient outcome: aneurysmal sub-arachnoid hemorrhage (aSAH). Our primary aim was identifying the prevalence of BAV in aSAH patients and compare it with that of a control group without aSAH. Secondary aims included neurologic outcomes of aSAH and anatomical characteristics of IA in BAV, and final disposition and outcome. Methods Patients (&amp;gt;18 years age) with an accurate diagnosis of aSAH and at least one echocardiogram in the system between January 2000-December 2019 were identified from a prospectively maintained registry of aSAH admissions. Controls, without a diagnosis of SAH ever, were age- and gender-matched (1:2 cases:controls). BAV prevalence was echocardiographically-confirmed in all patients. Severity of aSAH was categorized using the modified Fisher (mFisher) and the World Federation of Neurological Scale (WFNS). Results A total of 488 aSAH cases and 990 controls were included in the final analysis with confirmed aortic valve status. The prevalence of BAV in aSAH patients was 1.2% (6/488) vs 3.5% (35/990) in controls, p=0.01. None of the aSAH cases were noted to have CoA, whereas 3 (0.3%; p=0.12) control patients had presence of both BAV and CoA. Within aSAH patients, BAV+aSAH were noted to be younger than tricuspid aortic valve (TAV)+aSAH (56±11 vs 68±14; p=0.03). Aneurysm size was smaller in the BAV+aSAH group (5±2 vs 7±4 in TAV+aSAH; p=0.31). The severity of the bleed was lesser in BAV+aSAH than TAV+aSAH (mFisher grade &amp;gt;2 50% vs 74%; p=0.19, WFNS grade &amp;gt;3 17% vs 36%; p=0.43) which is clinically significant. The BAV+aSAH group had higher disposition rate to home (67% vs 39%; p=0.21) with no difference in in-hospital mortality rates (17% vs 18%; p=0.93). BAV had lower odds on univariate and multivariate analyses adjusted for age/sex/body mass index (OR 0.33 [CI 0.14–0.81]; p=0.01, OR 0.23 [CI 0.08–0.65]; p=0.01) of presenting with aSAH. Conclusion Within a large prospective registry with accurate aSAH diagnoses, we found 3-times lower prevalence of echocardiographically-confirmed BAV as compared to non-aSAH controls. We observed 3-times higher prevalence of BAV in controls than among aSAH cases. In addition, size of intracranial aneurysms was clinically smaller in BAV patients, and the severity of SAH was significantly lesser in BAV from a clinical standpoint. Our findings argue against routine surveillance for IA in BAV patients without CoA. Funding Acknowledgement Type of funding sources: None.

Detecting Aortic Valve Anomaly From Induced Murmurs: Insights From Computational Hemodynamic Models.

Patients who receive transcatheter aortic valve replacement are at risk for leaflet thrombosis-related complications, and can benefit from continuous, longitudinal monitoring of the prosthesis. Conventional angiography modalities are expensive, hospital-centric and either invasive or employ potentially nephrotoxic contrast agents, which preclude their routine use. Heart sounds have been long recognized to contain valuable information about individual valve function, but the skill of auscultation is in decline due to its heavy reliance on the physician’s proficiency leading to poor diagnostic repeatability. This subjectivity in diagnosis can be alleviated using machine learning techniques for anomaly detection. We present a computational and data-driven proof-of-concept analysis of a novel, auscultation-based technique for monitoring aortic valve, which is practical, non-invasive, and non-toxic. However, the underlying mechanisms leading to physiological and pathological heart sounds are not well-understood, which hinders development of such a technique. We first address this by performing direct numerical simulations of the complex interactions between turbulent blood flow in a canonical ascending aorta model and dynamic valve motion in 29 cases with healthy and stenotic valves. Using the turbulent pressure fluctuations on the aorta lumen boundary, we model the propagation of heart sounds, as elastic waves, through the patient’s thorax. The heart sound may be recorded on the epidermal surface using a stethoscope/phonocardiograph. This approach allows us to correlate instantaneous hemodynamic phenomena and valve motion with the acoustic response. From this dataset we extract “acoustic signatures” of healthy and stenotic valves based on principal components of the recorded sound. These signatures are used to train a linear discriminant classifier by maximizing correlation between recorded heart sounds and valve status. We demonstrate that this classifier is capable of accurate prospective detection of anomalous valve function and that the principal component-based signatures capture prominent audible features of heart sounds, which have been historically used by physicians for diagnosis. Further development of such technology can enable inexpensive, safe and patient-centric at-home monitoring, and can extend beyond transcatheter valves to surgical as well as native valves.

An Ensembled Anomaly Detector for Wafer Fault Detection

The production process of a wafer in the semiconductor industry consists of several phases such as a diffusion and associated defectivity test, parametric test, electrical wafer sort test, assembly and associated defectivity tests, final test, and burn-in. Among these, the fault detection phase is critical to maintain the low number and the impact of anomalies that eventually result in a yield loss. The understanding and discovery of the causes of yield detractors is a complex procedure of root-cause analysis. Many parameters are tracked for fault detection, including pressure, voltage, power, or valve status. In the majority of the cases, a fault is due to a combination of two or more parameters, whose values apparently stay within the designed and checked control limits. In this work, we propose an ensembled anomaly detector which combines together univariate and multivariate analyses of the fault detection tracked parameters. The ensemble is based on three proposed and compared balancing strategies. The experimental phase is conducted on two real datasets that have been gathered in the semiconductor industry and made publicly available. The experimental validation, also conducted to compare our proposal with other traditional anomaly detection techniques, is promising in detecting anomalies retaining high recall with a low number of false alarms.

Aortic Valve Status Detection for Heart Failure Patient with LVAD Using Deep Neural Networks

Purpose Continuous flow left ventricular assist devices (LVADs) can change aortic valve (AoV) biomechanics, which can result in reduced or lack of AoV opening. This can increase the risk of thrombotic events, gastrointestinal bleeding and denovo aortic incompetence. Therefore, detection of AoV opening and adjusting pump speed corresponding to the AoV status play a vital role for heart failure (HF) patients with LVADs. Although current assessment of AoV rely on supine resting echocardiography, this method cannot be applied outside of hospital. This study presents a deep neural network model for noninvasive detection of AoV opening for Heartware patients in different patient states. Methods 18 patients supported by a continuous flow HeartWare HVAD underwent transthoracic echocardiography for parasternal M‐mode images of the AoV. A continuous recording of 2 minutes at rest, immediately following by 1-minute Valsalva maneuver and then followed by leg‐raising exercise for 1-minute. A deep neural network model was trained and validated through 10-fold cross validation to detect AoV status based on the 200 samples (four seconds) of pump flow. The developed model performance was assessed using several performance metrics such as accuracy, sensitivity, specificity and receiver operating characteristic (ROC) curve area (AUC). Results The deep neural networks model was able to detect AoV opening/closing accurately across 10-fold cross validation based on the pump flow. The results show the accuracy of 86.59 %, sensitivity 81.39 %, specificity of 89.55 % and AUC of 0.928. However, including more patients on training and validation process of deep neural networks model can improve these performance metrics. Conclusion The deep neural network model would be useful for real-time detection of AoV opening which can be employed to adjust pump speed in HF patients to promote at least intermittent AV opening. Implementation of the proposed AoV detection in real-time mode could be integrated into a “smart pump” control system to automatically adjust LVADs speed to ensure a desired frequency of AV opening.