An Effective Machine-Learning Based Feature Extraction/Recognition Model for Fetal Heart Defect Detection from 2D Ultrasonic Imageries

Congenital heart defect,accounting for about 30%of congenital defects,is the most common one.Data shows that congenital heart defects have seriously affected the birth rate of healthy newborns.In Fetal andNeonatal Cardiology,medical imaging technology(2D ultrasonic,MRI)has been proved to be helpful to detect congenital defects of the fetal heart and assists sonographers in prenatal diagnosis.It is a highly complex task to recognize 2D fetal heart ultrasonic standard plane(FHUSP)manually.Compared withmanual identification,automatic identification through artificial intelligence can save a lot of time,ensure the efficiency of diagnosis,and improve the accuracy of diagnosis.In this study,a feature extraction method based on texture features(Local Binary Pattern LBP and Histogram of Oriented Gradient HOG)and combined with Bag of Words(BOW)model is carried out,and then feature fusion is performed.Finally,it adopts Support VectorMachine(SVM)to realize automatic recognition and classification of FHUSP.The data includes 788 standard plane data sets and 448 normal and abnormal plane data sets.Compared with some other methods and the single method model,the classification accuracy of our model has been obviously improved,with the highest accuracy reaching 87.35%.Similarly,we also verify the performance of the model in normal and abnormal planes,and the average accuracy in classifying abnormal and normal planes is 84.92%.The experimental results show that thismethod can effectively classify and predict different FHUSP and can provide certain assistance for sonographers to diagnose fetal congenital heart disease.

Study of ultrasonic phased array inspection imaging technology for NDT

A research about the ultrasonic phased array imaging principle from A-scan signal to B-scan image for non-destructive testing （NDT） was conducted in this paper, the ultrasonic phased array inspection imaging system used in industrial field was developed and the experiment was performed on the steel testing block by the system with 64 elements, 5 MHz phased array transducer. Experimental results show that the flaws could be accurately detected and the flaws size could be estimated from the B-scan images, and the B-scan images could clearly show the location of the flaws, but the quality of B-scan images needs to be enhanced by digital signal processing and controlling dynamic focusing for improving the image resolution.

An application of ultrasonic phased array imaging in electron beam welding inspection

The basic principle and features of ultrasonic phased array imaging are discussed in this paper. Through the ultrasonic phased array technology, the electron beam welding defects and frozen keyholes characterization and imaging were realized. The ultrasonic phased array technology can detect kinds of defects in electron beam welding (EBW) quickly and easily.

Simplified Matrix Focusing Imaging Algorithm for Ultrasonic Nondestructive Testing

Full matrix focusing method of ultrasonic phased array has been proved with advantages of good signal-to-noise ratio and imaging resolution in the field of Ultrasonic NDT.However,it is still suffering from the time-consuming data acquisition and processing.In order to solve the problem,two simplified matrix focusing methods are provided in the paper.One provided method is a triangular matrix focusing algorithm based on the principle of reciprocity for the multi-channel ultrasonic system.The other provided method is a trapezoidal matrix focusing algorithm based on the energy weight of the different channel to the focusing area.Time of data acquisition and computational is decreased with the provided simplified matrix focusing methods.In order to prove the validity of two provided algorithms,both side-drilled holes and oblique cracks are used for imaging experiments.The experimental results show that the imaging quality of the triangular matrix focusing algorithm is basically consistent to that of the full matrix focusing method.And imaging quality of the trapezoidal matrix focusing algorithm is slightly reduced with the amount of multi-channel data decreasing.Both data acquisition and computational efficiency using the triangular matrix focusing algorithm and the trapezoidal matrix focusing algorithm have been improved significantly compared with original full matrix focusing method.

Ultrasonic C-scanning imaging inspection of superplastic solid-state welded joint quality

Based on a large amount of dissection at welded interface and quantitative microscopic examination of welded rate, the suitable limit grey scale value was determined, and the welded rate of superplastic solid state welding interface of heterogeneous steel was systematically studied by means of self made ultrasonic C scanning imaging inspection system. The experimental results show: the welded state of superplastic solid state welding interface of heterogeneous steel can be conducted to be more accurately, reliably and quickly inspected by means of this system, and the ultrasonic testing results are good consistent with actual examination results of the interface defective distribution. Within the extent of the suitble welded rate,the welded rate in 40Cr/T10A superplastic welding process tested by this system is linear with its tensile strength of joint.

High-resolution bone microstructure imaging based on ultrasonic frequency-domain full-waveform inversion

The main challenge in bone ultrasound imaging is the large acoustic impedance contrast and sound velocity differences between the bone and surrounding soft tissue. It is difficult for conventional pulse-echo modalities to give accurate ultrasound images for irregular bone boundaries and microstructures using uniform sound velocity assumption rather than getting a prior knowledge of sound speed. To overcome these limitations, this paper proposed a frequency-domain fullwaveform inversion(FDFWI) algorithm for bone quantitative imaging utilizing ultrasonic computed tomography(USCT).The forward model was calculated in the frequency domain by solving the full-wave equation. The inverse problem was solved iteratively from low to high discrete frequency components via minimizing a cost function between the modeled and measured data. A quasi-Newton method called the limited-memory Broyden–Fletcher–Goldfarb–Shanno algorithm(L-BFGS) was utilized in the optimization process. Then, bone images were obtained based on the estimation of the velocity and density. The performance of the proposed method was verified by numerical examples, from tubular bone phantom to single distal fibula model, and finally with a distal tibia-fibula pair model. Compared with the high-resolution peripheral quantitative computed tomography(HR-p QCT), the proposed FDFWI can also clearly and accurately presented the wavelength scaled pores and trabeculae in bone images. The results proved that the FDFWI is capable of reconstructing high-resolution ultrasound bone images with sub-millimeter resolution. The parametric bone images may have the potential for the diagnosis of bone disease.

Backpropagation neural network method in data processing of ultrasonic imaging logging-while-drilling

The existing methods for extracting the arrival time and amplitude of ultrasonic echo cannot eff ectively avoid the local interference of ultrasonic signals while drilling,which leads to poor accuracy of the echo arrival time and amplitude extracted by an ultrasonic imaging logging-while-drilling tool.In this study,a demodulation algorithm is used to preprocess the ultrasonic simulation signals while drilling,and we design a backpropagation neural network model to fit the relationship between the waveform data and time and amplitude.An ultrasonic imaging logging model is established,and the finite element simulation software is used for forward modeling.The response under diff erent measurement conditions is simulated by changing the model parameters,which are used as the input layer of the neural network model;The ultrasonic echo signal is considered as a low-frequency signal modulated by a high-frequency carrier signal,and a low-pass fi lter is designed to remove the high-frequency signal and obtain the low-frequency envelope signal.Then the amplitude of the envelope signal and its corresponding time are extracted as an output layer of the neural network model.By comparing the application eff ects of the various training methods,we fi nd that the conjugate gradient descent method is the most suitable method for solving the neural network model.The performance of the neural network model is tested using 11 groups of simulation test data,which verify the eff ectiveness of the model and lay the foundation for further practical application.

An ultrasonic multi-wave focusing and imaging method for linear phased arrays

To overcome the inherent limits of traditional single wave imaging for nondestructive testing,the multi-wave focusing and imaging method is thoroughly studied.This method makes the compressional waves and shear waves focused in both emission and reception processes,which strengthens the focusing energy and improves the signal-to-noise ratio of received signals.A numerical model is developed to study the characteristics of a multi-wave focusing field.It is shown that the element width approaching 0.8 wavelengths of shear waves can keep a balance between the radiation energy of two waves,which can achieve a desirable multi-wave focusing performance.And an experiment using different imaging methods for a linear phased array is performed.It can be concluded that due to the combination of the propagation and reflection characteristics of two waves,the multi-wave focusing and imaging method can significantly improve the imaging distinguishability of defects and expand the available sweeping range to a sector of-650 to 65°.

Robust Image Watermarking Using LWT and Stochastic Gradient Firefly Algorithm

Watermarking of digital images is required in diversified applicationsranging from medical imaging to commercial images used over the web.Usually, the copyright information is embossed over the image in the form ofa logo at the corner or diagonal text in the background. However, this formof visible watermarking is not suitable for a large class of applications. In allsuch cases, a hidden watermark is embedded inside the original image as proofof ownership. A large number of techniques and algorithms are proposedby researchers for invisible watermarking. In this paper, we focus on issuesthat are critical for security aspects in the most common domains like digitalphotography copyrighting, online image stores, etc. The requirements of thisclass of application include robustness (resistance to attack), blindness (directextraction without original image), high embedding capacity, high Peak Signalto Noise Ratio (PSNR), and high Structural Similarity Matrix (SSIM). Mostof these requirements are conflicting, which means that an attempt to maximizeone requirement harms the other. In this paper, a blind type of imagewatermarking scheme is proposed using Lifting Wavelet Transform (LWT)as the baseline. Using this technique, custom binary watermarks in the formof a binary string can be embedded. Hu’s Invariant moments’ coefficientsare used as a key to extract the watermark. A Stochastic variant of theFirefly algorithm (FA) is used for the optimization of the technique. Undera prespecified size of embedding data, high PSNR and SSIM are obtainedusing the Stochastic Gradient variant of the Firefly technique. The simulationis done using Matrix Laboratory (MATLAB) tool and it is shown that theproposed technique outperforms the benchmark techniques of watermarkingconsidering PSNR and SSIM as quality metrics.

Research and Application of Log Defect Detection and Visualization System Based on Dry Coupling Ultrasonic Method

In order to optimize the wood internal quality detection and evaluation system and improve the comprehensive utilization rate of wood,this paper invented a set of log internal defect detection and visualization system by using the ultrasonic dry coupling agent method.The detection and visualization analysis of internal log defects were realized through log specimen test.The main conclusions show that the accuracy,reliability and practicability of the system for detecting the internal defects of log specimens have been effectively verified.The system can make the edge of the detected image smooth by interpolation algorithm,and the edge detection algorithm can be used to detect and reflect the location of internal defects of logs accurately.The content mentioned above has good application value for meeting the requirement of increasing demand for wood resources and improving the automation level of wood nondestructive testing instruments.

Internal Defects Detection Method of the Railway Track Based on Generalization Features Cluster Under Ultrasonic Images

There may be several internal defects in railway track work that have different shapes and distribution rules,and these defects affect the safety of high-speed trains.Establishing reliable detection models and methods for these internal defects remains a challenging task.To address this challenge,in this study,an intelligent detection method based on a generalization feature cluster is proposed for internal defects of railway tracks.First,the defects are classified and counted according to their shape and location features.Then,generalized features of the internal defects are extracted and formulated based on the maximum difference between different types of defects and the maximum tolerance among same defects’types.Finally,the extracted generalized features are expressed by function constraints,and formulated as generalization feature clusters to classify and identify internal defects in the railway track.Furthermore,to improve the detection reliability and speed,a reduced-dimension method of the generalization feature clusters is presented in this paper.Based on this reduced-dimension feature and strongly constrained generalized features,the K-means clustering algorithm is developed for defect clustering,and good clustering results are achieved.Regarding the defects in the rail head region,the clustering accuracy is over 95%,and the Davies-Bouldin index(DBI)index is negligible,which indicates the validation of the proposed generalization features with strong constraints.Experimental results prove that the accuracy of the proposed method based on generalization feature clusters is up to 97.55%,and the average detection time is 0.12 s/frame,which indicates that it performs well in adaptability,high accuracy,and detection speed under complex working environments.The proposed algorithm can effectively detect internal defects in railway tracks using an established generalization feature cluster model.

Automatic defect detection in ultrasonic TOFD D-scan data using image processing methods

In ultrasonic time of flight diffraction （TOFD） D-scan image, only a small fraction represents defects, whereas the majority is redundant. Because of the low contrast between defect and background image, it is difficult to manually interpret TOFD image. In addition, due to the nature of the weak amplitude of ultrasonic diffracted signals, the human factor introduces inconsistency into the interpretation. In order to automatically distinguish weld defects from the D-scan image, a defect detection method based on image processing technique is proposed. First, image pre-processing including clutter and noise suppression is conducted. Second, information entropy based image segmentation technique is employed to extract defects in the pre-processed image. At last, mathematical morphology based post-processing is carried out. The experimental results show that with the proposed method, TOFD can be used for automatic weld defect detection with satisfactory level of reliability.

Rapid ultrasonic C-scan on image test for spot welding based interpolation

In this paper, ultrasonic C-scan test of spot welds for stainless steel has been studied. It is concluded that large scanning step length contributes to high testing efficiency, however, the low-resolution C-scan image generated cannot be used to assess spot welding quality reliably. Based on bicubic image interpolation, the C-scan image in low resolution with the large step length 1 000 ~xm is subdivided and reconstructed. By this means, the C-scan image resolution is greatly enhanced and testing results obtained are satisfactory, realizing rapid assessment of spot welds. The results of rapid ultrasonic C-scan test fit the actual metallographic measured value well. Mean value of normal distribution of error statistics is O. 006 67, and the standard deviation is O. 087 11. Rapid ultrasonic C-scan test based on image interpolation is of high accuracy and excellent stability.

Ultrasonic image restoration based on support vector machine for surfacing interface testing

In order to restore the degraded ultrasonic C-scan image for testing surfacing inteoface, a method based on support vector regression （SVR） network is proposed. By using the image of a simulating defect, the network is trained and a mapping relationship between the degraded and restored image is founded. The degraded C-scan image of Cu-Steel surfacing inteoface is processed by the trained network and improved image is obtained. The result shows that the method can effectively suppress the noise and deblur the defect edge in the image, and provide technique support for quality and reliability evaluation of the surfacing weld.

Regardless of the Puncture Angle, a Tuohy Needle with Sand-Blasted Steel at the Tip of the Bevel Is a Valid Needle on the Ultrasonic Images

Background: A high ability to visualize the needle on the ultrasonic images is necessary to perform the ultrasound-guided nerve block safely. The Rafa Tuohy needleR (Vygon, Paris, France, Rafa) is a non-stimulating Tuohy needle with sand-blasted steel at the tip of the bevel. We examined the degree to which the Rafa enhanced the visibility of ultrasonic images compared with the non-coated Tuohy needle. Methods: We punctured the Blue Phantom. The dimensions of both the Rafa and the non-coated Tuohy needles were 18 G × 80 mm. The puncture angle is 30 degrees and 45 degrees from the Blue Phantom. We measured the intensity of the tip of the bevel at a depth of 0.5, 1.0, 1.5 and 2.0 cm for the puncture angle of 30 degrees, and 0.5, 1.0, 1.5, 2.0, 2.5 and 3.0 cm for the puncture angle of 45 degrees. Six anesthesiologists with more than seven years of experience performed three punctures using each needle. Results: As an outcome, we concluded that at a puncture angle of both 30 degrees and 45 degrees, the intensity of the non-coated Tuohy needle decreased with the depth. On the other hand, at an angle of 30 degrees, the intensity of the Rafa needle did not decrease, and at an angle of 45 degrees the intensity only decreased very slightly as the depth increased. Conclusions: The Tuohy needle with sand-blasted steel at the tip of the bevel provided greater visualization than the non-coated Tuohy needle on the ultrasound images.

Analysis and Management System of Digital Ultrasonic Image

This paper presents the analysis and management system of digital ultrasonic image. The system can manage medical ultrasonic image by collecting, saving and transferring, and realize that section offices of ultrasonic image in hospital network manage. The system use network technology in transferring image between ultrasonic equipments to share patient data in ultrasonic equipments. And doctors can input patient diagnostic report,saved by text file and case history, digitally managed. The system can be realized by Visual C++ which make windows applied. The system can be brought forward because PACS prevail with various hospitals,but PACS is expensive. In view of this status, we put forward to the analysis and management system of digital ultrasonic image,which is similar to PACS.

Finite element modeling of heating phenomena of cracks excited by high-intensity ultrasonic pulses

A three-dimensional thermo-mechanical coupled finite element model is built up to simulate the phenomena of dynamical contact and frictional heating of crack faces when the plate containing the crack is excited by high-intensity ultrasonic pulses. In the finite element model, the high-power ultrasonic transducer is modeled by using a piezoelectric thermal-analogy method, and the dynamical interaction between both crack faces is modeled using a contact-impact theory. In the simulations, the frictional heating taking place at the crack faces is quantitatively calculated by using finite element thermal-structural coupling analysis, especially, the influences of acoustic chaos to plate vibration and crack heating are calculated and analysed in detail. Meanwhile, the related ultrasonic infrared images are also obtained experimentally, and the theoretical simulation results are in agreement with that of the experiments. The results show that, by using the theoretical method, a good simulation of dynamic interaction and friction heating process of the crack faces under non-chaotic or chaotic sound excitation can be obtained.

Ultrasonic diagnosis of biliary atresia: A retrospective analysis of 20 patients

AIM: To investigate the clinical value of ultrasono-graphic diagnosis of biliary atresia (BA), a retrospective analysis of the sonogram of 20 children with BA was undertaken. METHODS: Ultrasonography (US) was performed in 20 neonates and infants with BA, which was confirmed with cholangiography by operation or abdominoscopy. The presence of triangular cord, the size and echo of liver, the changes in empty stomach gallbladder and postprandial gallbladder were observed and recorded. RESULTS: The triangular cord could be observed at the porta hepatis (thickness: 0.3-0.6 cm) in 10 cases. Smaller triangular cord (0.2-0.26 cm) can be observed in 3 cases. The gallbladder was not observed in 2 cases, and 1 case showed a streak gallbladder without capsular space. The gallbladders of 15 cases were flat and small. The gallbladders of 2 cases were of normal size and appearance, however, there was no postprandial contraction. The livers of all cases showed hepatomegaly and hetreogeneous echogenicity. Statistical analysis was performed to compare the hepatomegaly and hetreogeneous echogenicity and the stage of hepatic fibrosis. CONCLUSION: The presence of the triangular cord atthe porta hepatis is specific. However, it is not the only diagnostic criterion, since flat and small gallbladder and poor contraction are also of important diagnostic and differential diagnostic significance. The degree of hepatomegaly and hetreogeneous echogenicity is proportional with liver fibrosis, and able to indicate the duration of course and prognosis.

Ultrasonic Scattered Field Distribution of One and Two Cylindrical Solids with Phased Array Technique

The scattered fields of plane waves in a solid from a cylinder or sphere are critical in determining its acoustic characteristics as well as in engineering applications. This paper investigates the scattered field distributions of different incident waves created by elastic cylinders embedded in an elastic isotropic medium. Scattered waves, including longitudinal and transverse waves both inside and outside the cylinder, are described with specific modalities under an incident plane wave. A model with a scatterer embedded in a structural steel matrix and filled with aluminum is developed for comparison with the theoretical solution. The frequency of the plane wave ranged from 235 kHz to 2348 kHz, which corresponds to scaling factors from 0.5 to 5. Scattered field distributions in matrix materials blocked by an elastic cylindrical solid have been obtained by simulation or calculated using existing parameters. The simulation results are in good agreement with the theoretical solution, which supports the correctness of the simulation analysis. Furthermore, ultrasonic phased arrays are used to study scattered fields by changing the characteristics of the incident wave. On this foundation, a partial preliminary study of the scattered field distribution of double cylinders in a solid has been carried out, and the scattered field distribution at a given distance has been found to exhibit particular behaviors at different moments. Further studies on directivities and scattered fields are expected to improve the quantification of scattered images in isotropic solid materials by the phased array technique.

Use of intravascular imaging in managing coronary artery disease

For many years,coronary angiography has been considered "the gold standard" for evaluating patients with coronary artery disease. However,angiography only provides a planar two-dimensional silhouette of the lumen and is unsuitable for the precise assessment of atherosclerosis. With the introduction of intravascular imaging,direct visualization of the arterial wall is now feasible. Intravascular imaging modalities extend diagnostic information,thereby enabling more precise evaluation of plaque burden and vessel remodeling. Of all technologies,intravascular ultrasound(IVUS) is the most mature and widely used intravascular imaging technique. Optical coherence tomography(OCT) is an evolving technology that has the highest spatial resolution of existing imaging methods,and it is becoming increasingly widespread. These methods are useful tools for planning interventional strategies and optimizing stent deployment,particularly when stenting complex lesions. We strongly support the mandatory use of IVUS for left main percutaneous coronary intervention(PCI). In addition,it can be used to evaluate vascularresponses,including neointimal growth and strut apposition,during follow-ups. Adequately powered randomized trials are needed to support IVUS or OCT use in routine clinical practice and to answer whether OCT is superior to IVUS in reducing adverse events when used to guide PCI. The current perception and adoption of innovative interventional devices,such as bioabsorbable scaffolds,will increase the need for intravascular imaging in the future.