Least-squares estimation of imaging parameters for an ultrasonic array using known geometric image features

A Direct Wavepath-based Element Localization Algorithm to Enable Flexible Ultrasound Array Imaging

An algorithm is developed for determining the element locations of a flexible ultrasonic array when applied to a surface of unknown geometry. The algorithm forms a dataset of traveltimes from the direct wavepaths (i.e. rays) between transmitters and receivers, which serves as the input to an optimization scheme that iterates on the array element locations until an objective function is minimized. Once, the relative array locations have been determined, they are used as an input to a phased array ultrasound imaging algorithm. In this study, the total focusing method with full matrix capture is used as a testbed code to demonstrate the benefits of the relative array element localization algorithm. The algorithm is verified by simulation and experimentation.

Flexible Ultrasound Transducer With Embedded Optical Shape Sensing Fiber for Biomedical Imaging Applications

Flexible ultrasound transducers (FUTs), capable of conforming to irregular surfaces, have become a research hotspot in the field of medical imaging. With these transducers, high-quality ultrasound images can be obtained only if strict design criteria are fulfilled. Moreover, the relative positions of array elements must be determined, which are important for ultrasound beamforming and image reconstruction. These two major characteristics present great challenges to the design and fabrication of FUTs compared to that for traditional rigid probes. In this study, an optical shape-sensing fiber was embedded into a 128-element flexible linear array transducer to acquire the real-time relative positions of array elements to produce high-quality ultrasound images. Minimum concave and convex bend diameters of approximately 20 and 25 mm, respectively, were achieved. The transducer was flexed 2000 times, and yet no obvious damage was observed. Stable electrical and acoustic responses confirmed its mechanical integrity. The developed FUT exhibited an average center frequency of 6.35 MHz, and average -6-dB bandwidth of 69.2%. The array profile and element positions measured by the optic shape-sensing system were instantly transferred to the imaging system. Phantom experiments for both spatial resolution and contrast-to-noise ratio proved that FUTs can maintain satisfactory imaging capability despite bending to sophisticated geometries. Finally, color Doppler images and Doppler spectra of the peripheral arteries of healthy volunteers were obtained in real time.

Studies of Angular Resolution for Acoustic Arc Arrays

Currently, phased arrays are increasingly used in ultrasonic nondestructive testing. One of the most important parameters of ultrasonic nondestructive testing with the application of phased arrays is the angular resolution. This paper presents the results of studies of the angular resolution of concave and convex acoustic arrays in ultrasonic testing with the application of the total focusing method. Computer modeling of concave and convex acoustic arrays consisting of 16, 32 and 64 elements with distances between elements of 0.5 and 1 mm and arc radii of 30 and 60 mm have been performed. The results obtained by computer modeling were confirmed via in situ experiments.

Towards Explainable Augmented Intelligence (AI) for Crack Characterization

Crack characterization is one of the central tasks of NDT&E (the Non-destructive Testing and Evaluation) of industrial components and structures. These days data necessary for carrying out this task are often collected using ultrasonic phased arrays. Many ultrasonic phased array inspections are automated but interpretation of the data they produce is not. This paper offers an approach to designing an explainable AI (Augmented Intelligence) to meet this challenge. It describes a C code called AutoNDE, which comprises a signal-processing module based on a modified total focusing method that creates a sequence of two-dimensional images of an evaluated specimen; an image-processing module, which filters and enhances these images; and an explainable AI module—a decision tree, which selects images of possible cracks, groups those of them that appear to represent the same crack and produces for each group a possible inspection report for perusal by a human inspector. AutoNDE has been trained on 16 datasets collected in a laboratory by imaging steel specimens with large smooth planar notches, both embedded and surface-breaking. It has been tested on two other similar datasets. The paper presents results of this training and testing and describes in detail an approach to dealing with the main source of error in ultrasonic data—undulations in the specimens’ surfaces.

Flexible ultrasonic array for breast-cancer diagnosis based on a self-shape-estimation algorithm

Breast cancer is a very common malignant tumour that typically occurs in women aged 35-70 years (accounting for 85% of patients). Recently, it has been appearing in younger women as well. Traditional ultrasonic transducers usually use a fixed array, which avoids the radiation from mammography, has a low cost, and can be used for repeated testing. This substantially benefits the clinical diagnosis of breast cancer. However, the fixed transducer-array diagnosis process exerts considerable pressure on the human body, which can easily cause mass displacement or unnecessary pain. Therefore, ultrasound breast cancer diagnosis without compression has attracted attention. In this study, we used a flexible ultrasonic array to record the ultrasound information of the mass, and proposed a mathematical model suitable for breast-cancer diagnosis. Then, we used a self-shape-estimation algorithm to obtain a two-dimensional (2D) ultrasound image of the breast cancer. The algorithm was tested with simulated and experimental array data, and its performance was evaluated according to the tumour location. The surface-shape error obtained through the numerical simulation was less than 0.8 mm, and the deviation in the estimated mass position was less than 1.24 mm. The tumour location was also obtained experimentally in a breast-cancer model. Therefore, the method proposed in this paper can realize ultrasound diagnoses and represents a new diagnostic tool for breast cancer.

Efficient immersion imaging of components with nonplanar surfaces

Ultrasonic array inspection of a component with a nonplanar surface can be achieved in immersion using a liquid layer to couple ultrasonic waves from an array probe into a solid structure. This paper presents an efficient way to compute the appropriate element time delays in immersion without compromising the measurement accuracy. In the proposed imaging process, the surface geometry is first measured ultrasonically by forming an image of the component surface in the couplant. This leads to a set of discrete points that define the surface profile of the component. The propagation time from an array element to a point in the component is then determined by a grid search of candidate ray-paths through each surface point to identify the one that yields the shortest traveling time. Propagation times in the component are first generated on a coarse mesh of points and then these values are linearly interpolated to find the propagation time to each image pixel. The computed propagation times are finally used to reconstruct an image of the component interior. An analytical model is developed to determine a relationship between estimated propagation time errors and their effect on the array inspection in terms of signal amplitude from a reflector. For nominally normal incidence inspection of a metallic component with a minimum surface radius of 30 wavelengths immersed in water, it is found that the surface of the component can be adequately described by points spaced by one wavelength and that delays can be computed on a coarse grid of points spaced at 3 wavelengths. With these parameters, the reduction in amplitude of a point target in the component is shown to be less than 1 dB.