Lung ultrasound is an increasingly utilized non-invasive imaging modality for assessing lung condition but interpreting it can be challenging and depends on the operator’s experience. To address these challenges, this work proposes an approach that combines artificial intelligence (AI) with feature-based signal processing algorithms. We introduce a specialized deep learning model designed and trained to facilitate the analysis and interpretation of lung ultrasound images by automating the detection and location of pulmonary features, including the pleura, A-lines, B-lines, and consolidations. Employing Convolutional Neural Networks (CNNs) trained on a semi-automatically annotated dataset, the model delineates these pulmonary patterns with the objective of enhancing diagnostic precision. Real-time post-processing algorithms further refine prediction accuracy by reducing false-positives and false-negatives, augmenting interpretational clarity and obtaining a final processing rate of up to 20 frames per second with accuracy levels of 89% for consolidation, 92% for B-lines, 66% for A-lines, and 92% for detecting normal lungs compared with an expert opinion.

Plane Wave Imaging (PWI) has been recently proposed for fast ultrasound inspections in the Non-Destructive-Testing (NDT) field. By using a single (or a reduced number) of plane wave emissions and parallel beamforming in reception, frame rates of hundreds to thousands of images per second can be achieved without significant image quality losses with regard to the Total Focusing Method (TFM) or Phased Array (PA). This work addresses the problem of applying PWI in the presence of arbitrarily shaped interfaces, which is a common problem in NDT. First, the mathematical formulation for generating a plane wave inside a component of arbitrary geometry is given, and the characteristics of the resultant acoustic field are analyzed by simulation, showing plane wavefronts with non-uniform amplitude. Then, an imaging strategy is proposed, accounting for this amplitude effect. Finally, the proposed method is experimentally validated, and its application limits are discussed.

Ultrasonic non-destructive testing (UT) is a proficient method for detecting damage in composite materials; however, conventional manual testing procedures are time-consuming and labor-intensive. We propose a semi-automated defect segmentation methodology employing a convolutional neural network (CNN) on 3D ultrasonic data, facilitated by the fusion of X-ray computed tomography (XCT) and Phased-Array Ultrasonic Testing (PAUT) data. This approach offers the ability to develop supervised datasets for cases where UT techniques inadequately assess defects and enables the creation of models with genuine defects rather than artificially introduced ones. During the training process, we recommend processing the 3D volumes as a sequence of 2D slices derived from each technique. Our methodology was applied to segment porosity, a common defect in composite materials, for which characteristics such as void size and shape remain immeasurable via UT. Precision, recall, F1 score, and Intersection over Union (IoU) metrics were used in the evaluation. The results of the evaluation show that the following challenges have to be faced for improvement: (i) achieving accurate 3D registration, (ii) discovering suitable similar keypoints for XCT and UT data registration, (iii) differentiating ultrasonic echoes originating from porosity versus those related to noise or microstructural features (interfaces, resin pockets, fibers, etc.), and, (iv) single out defect echoes located near the edges of the component. In fact, an average F1 score of 0.66 and IoU of 0.5 were obtained.

This paper describes the design and implementation of an ultrasonic non-contact air-coupled technique (UNCACT) using antisymmetric Lamb waves (ALW) for NDT assessments in novel composite sandwich plates of a car body shell. This technique is complemented with a C-Scan image implementation using guided waves. The finite element method (FEM) was developed using Comsol 6.1 for the interpretation of the several wave modes presented in the experiments, including the ALW mode. This FEM model is indispensable for the correct interpretation of the received signals and contributes to a better implementation of this technology. This is a novel contribution building upon previously reported work. Additionally, the phase velocity method (PVM) was applied for the verification of the ALW mode in the portion of the RF signal necessary for the C-Scan image.

Lung ultrasound is an increasingly utilized non-invasive imaging modality for assessing the lung condition, but interpreting it can be challenging and depends on the operator experience. To address these challenges, this work proposes an approach that combines artificial intelligence (AI) with feature-based signal processing algorithms. We introduce a specialized deep learning model designed and trained to facilitate the analysis and interpretation of lung ultrasound images, by automating the detection and location of pulmonary features, including the pleura, A-lines, B-lines and consolidations. Employing Convolutional Neural Networks (CNNs) trained on a semi-automatically annotated dataset, the model delineates these pulmonary patterns, with the objective of enhancing diagnostic precision. Real-time post-processing algorithms further refine prediction accuracy by reducing false-positives and false-negatives, augmenting interpretational clarity and obtaining a final processing rate of up to 20 frames per second withl accuracies of 89% for consolidation, 92% for B-lines and 66% in case of A-lines compared with an expert opinion.