Early thrombus detection in the extracorporeal membrane oxygenation circuit by noninvasive real-time ultrasonic sensors

Thrombus formation in extracorporeal membrane oxygenation (ECMO) remains a major concern as it can lead to fatal outcomes. To the best of our knowledge, there is no standard non-invasive method for quantitatively measuring thrombi. This study's purpose was to verify thrombus detection in an ECMO circuit using novel, non-invasive ultrasonic sensors in real-time, utilizing the fact that the ultrasonic velocity in a thrombus is known to be higher than that in the blood. Ultrasonic sensors with a customized chamber, an ultrasonic pulse-receiver, and a digital storage oscilloscope (DSO) were used to set up the measuring unit. The customized chamber was connected to an ECMO circuit primed with porcine blood. Thrombi formed from static porcine blood were placed in the circuit and ultrasonic signals were extracted from the oscilloscope at various ECMO flow rates of 1-4 L/min. The ultrasonic signal changes were successfully detected at each flow rate on the DSO. The ultrasonic pulse signal shifted leftward when a thrombus passed between the two ultrasonic sensors and was easily detected on the DSO screen. This novel real-time non-invasive thrombus detection method may enable the early detection of floating thrombi in the ECMO system and early management of ECMO thrombi.

Deep learning techniques to detect rail indications from ultrasonic data for automated rail monitoring and maintenance

The increasing number of passengers and services using railways and the corresponding increase in rail use has caused the acceleration of rail wear and surface defects which makes rail defect identification an important issue for rail maintenance and monitoring to ensure safe and efficient operation. Traditional visual inspection methods for identifying rail defects are time-consuming, less accurate, and associated with human errors. Deep learning has been used to improve railway maintenance and monitoring tasks. This study aims to develop a structured model for detecting railway artifacts and defects by comparing different deep-learning models using ultrasonic image data. This research showed whether it is practical to identify rail indications using image classification and object detection techniques from ultrasonic data and which model performs better among the above-mentioned methods. The methodology includes data processing, labeling, and using different conventional neural networks to develop the model for both image classification and object detection. The results of CNNs for image classification, and YOLOv5 for object detection show 98%, and 99% accuracy respectively. These models can identify rail artifacts efficiently and accurately in real-life scenarios, which can improve automated railway infrastructure monitoring and maintenance.

Improved water resource management framework for water sustainability and security

The water resource is an essential field of economic growth, social progress, and environmental integrity. A novel solution is offered to meet water needs, distribution, and IoT-based quality management requirements. With technological growth, this paper presents an IoT-enabled Water Resource Management and Distribution Monitoring System (IWRM-DMS) using sensors, gauge meters, flow meters, ultrasonic sensors, motors to implement in rural cities. Thus, research proposes that the IWRM-DMS establish the rural demand for water and the water supply system to minimize water demand. The system proposed includes different sensors, such as the water flow sensor, the pH sensor, the water pressure valve, the flow meters, and ultrasound sensors. This water system has been developed, which addresses the demand for domestic water in the village. Machine Intelligence has been designed for demand prediction in the decision support system. The simulation results confirm the applicability of the proposed framework in real-time environments. The proposed IWRM-DMS has been proposed to analyse the water quality to ensure water distribution in a rural area to achieve less MAPE (21.41%) and RMSE(15.12%), improve efficiency (96.93%), Reliability (98.24%), enhance prediction (95.29%)), the overall performance (97.34%), moisture content ratio (7.4%), cost-effectiveness ratio (95.7%) when compared to other popular methods.

Transfer learning for process monitoring using reflection-mode ultrasonic sensing

The fourth industrial revolution is set to integrate entire manufacturing processes using industrial digital technologies such as the Internet of Things, Cloud Computing, and machine learning to improve process productivity, efficiency, and sustainability. Sensors collect the real-time data required to optimise manufacturing processes and are therefore a key technology in this transformation. Ultrasonic sensors have benefits of being low-cost, in-line, non-invasive, and able to operate in opaque systems. Supervised machine learning models can correlate ultrasonic sensor data to useful information about the manufacturing materials and processes. However, this requires a reference measurement of the process material to label each data point for model training. Labelled data is often difficult to obtain in factory environments, and so a method of training models without this is desirable. This work compares two domain adaptation methods to transfer models across processes, so that no labelled data is required to accurately monitor a target process. The two method compared are a Single Feature transfer learning approach and Transfer Component Analysis using three features. Ultrasonic waveforms are unique to the sensor used, attachment procedure, and contact pressure. Therefore, only a small number of transferable features are investigated. Two industrially relevant processes were used as case studies: mixing and cleaning of fouling in pipes. A reflection-mode ultrasonic sensing technique was used, which monitors the sound wave reflected from the interface between the vessel wall and process material. Overall, the Single Feature method produced the highest prediction accuracies: up to 96.0% and 98.4% to classify the completion of mixing and cleaning, respectively; and R² values of up to 0.947 and 0.999 to predict the time remaining until completion. These results highlight the potential of combining ultrasonic measurements with transfer learning techniques to monitor industrial processes. Although, further work is required to study various effects such as changing sensor location between source and target domains.

Generation of sub-microsecond quasi-unipolar pressure pulses

We present a method to generate sub-microsecond quasi-unipolar pressure pulses. Our approach is based on the laser irradiation of a thin copper wire submerged in water. The acoustic waveforms were recorded using two different, well characterized, wideband detection techniques: piezoelectric and optical interferometry. The results show that the irradiated target behaves as an omnidirectional source. Moreover, the peak pulse pressure linearly depends on the laser fluence and the source size. From the results, we propose an empirical equation for the spatial and temporal profile of the pressure pulse. The method has several advantages: ease of implementation, high repeatability, wide ultrasonic bandwidth and quasi-unipolar time profile. These features lead to potential applications of this acoustic source in ultrasonic characterization such as transducer systems, materials or passive devices.

Performance optimization of high-order Lamb wave sensors based on silicon carbide substrates

Silicon carbide (SiC), as a new type of material for substrates in micro-electromechanical system (MEMS), was given high consideration in virtue of the properties of high acoustic velocity, low loss, chemical resistance, and etc. In this work, five performance parameters, which are electromechanical coupling coefficients, mass sensitivities, conductivity sensitivities, insert losses and minimum detectable masses, are theoretically investigated in Lamb wave chemical sensors for gas sensing based on SiC substrates. It is presented that higher performance can be achieved based on high-order modes other than fundamental modes, and the abovementioned five parameters can be simultaneously optimized. Then, according to the optimized operating conditions, operating parameters of the SiC-based high-order Lamb wave sensors are designed, which can be easily realized in MEMS technology. Finally, it is demonstrates that the SiC-based sensor exhibits better performance than that of the sensor with a conventional silicon substrate.

Optimization of electric bicycle for youths with disabilities

Cerebral palsy is a group of neurodevelopmental disorders that affect a person’s ability to move and to maintain balance and posture. People with cerebral palsy have also perception and space orientation deficits so that special assistance devices should be developed to compensate these handicaps. The objective was to optimize an adapted electric bicycle (E-bike) for youths with neurodevelopmental disorders. An adapted E-bike was provided with ultrasonic sensors that measure distances to objects. If the distance to other objects reduces, an acoustic signal is sent. Additionally, a self-created force plate was fixed on the pedal to evaluate the muscle performances during biking. An experiment with the ultrasound warning system confirmed that acoustic feedback was helpful in avoiding obstacles. The measurement of the blood pressure, the heart frequency and the pedaling force during biking approved that the training condition of the test person can be registered and enables tuning the power of the electric motor to individual requirements. The results demonstrate that an adapted E-bike can be improved to provide better space orientation for people with perceptual disorders and to measure training conditions of patients. Moreover, these enable individual adjustment of the electric motor power to optimize comfort and therapy effect.

An acoustic transmission sensor for the longitudinal viscosity of fluids

Physical fluid parameters like viscosity, mass density and sound velocity can be determined utilizing ultrasonic sensors. We introduce the concept of a recently devised transmission based sensor utilizing pressure waves to determine the longitudinal viscosity, bulk viscosity, and second coefficient of viscosity of a sample fluid in a test chamber. A model is presented which allows determining these parameters from measurement values by means of a fit. The setup is particularly suited for liquids featuring higher viscosities for which measurement data are scarcely available to date. The setup can also be used to estimate the sound velocity in a simple manner from the phase of the transfer function.

Sensing the characteristic acoustic impedance of a fluid utilizing acoustic pressure waves

Ultrasonic sensors can be used to determine physical fluid parameters like viscosity, density, and speed of sound. In this contribution, we present the concept for an integrated sensor utilizing pressure waves to sense the characteristic acoustic impedance of a fluid. We note that the basic setup generally allows to determine the longitudinal viscosity and the speed of sound if it is operated in a resonant mode as will be discussed elsewhere. In this contribution, we particularly focus on a modified setup where interferences are suppressed by introducing a wedge reflector. This enables sensing of the liquid's characteristic acoustic impedance, which can serve as parameter in condition monitoring applications. We present a device model, experimental results and their evaluation.