Nowadays, GPUs (Graphics Processing Units) serve as work-horses for processing massive amount of data and to accelerate general-purpose scientific and engineering computing. The main goal of the training is to get familiar with the GPU/parallel programming and apply it to ultrasound signal processing. The short-course is going to be practically oriented with a 50/50 split between the lectures and exercises. We are planning to leverage a common knowledge of the basic ultrasound processing methods and show how to translate them into working parallel algorithms. The workshop will target both low-level Nvidia CUDA GPU programming and high-level Python tools. This blend of development tools enable fast prototyping of new processing methods and later migration to a high-performance native GPU implementation. During the exercises, the Participants will implement and test their algorithms on ready to use RF datasets, as well as have an opportunity to run them on an ultrasound research system equipped with GPU.

This course covers capacitive micromachined ultrasonic transducers (CMUTs) and their applications. It begins with a background discussion of previous implementations, followed by a detailed explanation of the parallel plate capacitor transducer and its electrical circuit model. An analytical model that better represents the realizable membrane of a CMUT is developed, with motivation for a more sophisticated finite element model. CMUTs for airborne and immersion applications are compared and contrasted, along with two acoustic cross-talk modeling techniques. The two main CMUT fabrication techniques are explained and compared. Device characterization covers electrical, mechanical, and ultrasonic measurements, including low-noise front-end electronics design and integration. The course concludes with two detailed design examples for chemical/biological sensing and medical imaging applications.

Deep learning is a powerful tool that has quickly spread across many domains, including medical imaging. It is gaining attention in ultrasound imaging, and this course will explore the opportunities it brings. We will start with a brief introduction to deep learning and discuss its impact on medical imaging in general. Then, we will outline the opportunities for ultrasound imaging and the challenges that deep learning can address. Next, we will cover the fundamentals of deep learning, including optimization aspects and designing effective neural network architectures for ultrasound imaging. We will place a particular emphasis on model-based deep learning methods. Lastly, we will discuss neural networks for front-end receive processing and the power of end-to-end optimization of entire signal processing chains in ultrasound imaging.

Piezoelectric acoustic wave devices require high precision and accuracy for use as frequency standards and detection elements. Accurate 2-D or 3-D models are needed to maintain frequency performance through precise designs, manufacturing, and operations. These models should include nonlinear effects such as temperature sensitivity, applied forces, harmonic generation, and intermodulation. The course will cover linear and nonlinear finite element modeling of resonators, including material properties and deformations. The nonlinear behavior of quartz resonators and frequency response modeling for force frequency effects, harmonic generation, and intermodulation of BAW and SAW resonators will be discussed if time permits.

This course provides an introduction to the use of ultrasound in therapy, covering both bio-effects and mechanical effects. It includes a review of ultrasound generation and absorption leading to heat creation, as well as non-thermal therapy approaches such as lithotripsy, histotripsy, non-thermal ablation, and targeted drug delivery. The course also discusses current therapy ultrasound systems and their use in clinical practice, along with potential future directions and impacts. Emphasis is placed on technological issues and system architecture constraints, and examples of clinical study results are reviewed.

This course will cover the fundamental principles of acoustic wave theory, imaging, and inversion. You will learn about the acoustic field equations, pressure and velocity wave fields, wave equations for linear and non-linear acoustics, and solution methods for modeling acoustic wave fields in heterogeneous media. The course will also cover concepts such as Kirchhoff integrals, Rayleigh I and II integrals, and evanescent waves. You will gain an understanding of imaging and (non-linear) inversion for quantitative imaging. Overall, this course is designed to provide you with a comprehensive understanding of acoustic wave theory and its applications in imaging and inversion.

This short course is about laser ultrasonics, a non-contact and non-invasive approach for conducting ultrasonic measurements using optical generation and the detection of acoustic waves. The course covers the physical principles and mechanisms of optical generation of surface acoustic waves (SAWs) and established and emergent experimental techniques for laser generation and detection of SAWs. Recent developments such as using extreme ultraviolet light to generate SAWs up to 50 GHz in frequency are also discussed. The use of laser-generated SAWs for studying fundamental wave propagation phenomena including surface phononic crystals and metamaterials, as well as material science applications such as the characterization of elastic properties of thin films and testing material strength under ultrahigh strain rates, are explained. Finally, potential industrial applications for metrology and process control are discussed.

This course covers super-resolution ultrasound imaging, which can visualize structures smaller than the classical limit, such as blood vessels in deep tissue. It explores bypassing the diffraction limit with microbubbles, focusing on ultrasound localization microscopy. Applications in brain, tumor, kidney, liver, lymph node, and peripheral vessel imaging are detailed, with hands-on tips for conducting in vivo experiments on different animals and humans. The course also includes hands-on image processing of open data with provided algorithms and introduces deep learning-based microbubble localization and tracking methods, with practical experience in generating data for training and testing neural networks.

This short course covers the increasing importance of nonlinearity in ultrasonic waves in various fields. Nonlinear effects can lead to signal corruption in micro-acoustic devices and require countermeasures, while in non-destructive evaluation, nonlinearity is desired to detect defects at an earlier stage. The physical origin of classical and non-classical nonlinearity will be explained, along with a brief introduction to the theory of electro-elasticity. The course will focus on surface acoustic waves and cover dynamic nonlinear effects like harmonic generation, nonlinear mixing, nonlinear Cherenkov radiation, waveform evolution, and solitary waves. The perturbation theory approach will be demonstrated for some effects, along with other modeling tools like the finite element method. This course aims to provide a better understanding of nonlinearity for efficient use and design of countermeasures.

This short course is based on a recent review article (https://ieeexplore.ieee.org/abstract/document/9913943) and will present basic principles of hydrophone measurements, including mechanisms of action for various hydrophone designs, sensitivity, and directivity calibration procedures, practical considerations for performing measurements, signal processing methods to correct for both frequency- dependent sensitivity and spatial averaging across the hydrophone sensitive element, uncertainty in hydrophone measurements, special considerations for high-intensity therapeutic ultrasound, and advice for choosing an appropriate hydrophone for a particular measurement task. Recommendations will be made for information to be included in hydrophone measurement reporting. The instructors are world-leading hydrophone experts who are active in the development of International Electrotechnical Commission standards on hydrophones and collectively have authored over 50 papers concerning hydrophone methodology in peer-reviewed journals.

Acoustic wave resonators are becoming increasingly miniaturized for frequency control and sensing applications, creating a need for computer-aided design with strong formulation and modeling considerations. This course presents fundamental theories of acoustic wave devices, with emphasis on device structures, materials, and modes/frequencies. Simplifications are made for resonators with characteristic frequencies in operating modes. Numerical methods can be used to optimize actual resonator structures for better performance and accurate extraction of electrical parameters. The formulation is applied to common resonator types, including BAW, SAW, and FBAR. This course serves as an introduction to performance characteristics and provides the basis for further engineering solutions using tools such as finite element analysis.

In this tutorial I will present architectures based on piezoMEMS technology to demonstrate stress-optical modulation and tuning of silicon nitride and silicon photonic integrated circuits. We will define fundamental performance metrics and compare various monolithic and heterogeneous optomechanical systems. In the second part of the course, I will introduce new applications enabled by optomechanics including acousto-optic modulators, inertial sensors, magnetic-free optical isolators, and fast tunable lasers for LIDAR and microcombs. I will then close with a discussion of applications of ultrasonic piezoMEMS technology and frequency control not only of photons (flying qubits) but also color centers in diamond and silicon carbide (atom-like defects).

This short course provides an overview of Analog Front End (AFE) electronics and its interaction with passive transducers, as well as the integration of AFE with in-probe electronics. It begins with a primer on basic electronics principles such as Impedance Matching, Cable Selection, Transmit Circuits, Receive AFE, and ADC. The use of ASICs and the integration of ASICs with ultrasound transducer arrays are also discussed. Approaches to realizing in-probe transmit and receive circuitry, as well as channel-count reduction for high-element-count 3D probes, are introduced. The course then explores ultrasound system design in the context of passive probes with separate AFEs and active probes with in-probe electronics, covering data acquisition, storage, and transfer, as well as beamformer implementations and data post-processing for B-mode and Doppler modes. Finally, communication with active probes and the handling of pre-beamformed data are discussed.

In this short course, we present machine learning, signal processing algorithms, and high-performance computational system design for ultrasonic imaging, machine learning and data analysis applications. Several case studies will be presented including detecting defects in critical components in nuclear power plants, microstructure characterization and flaw detection in large-grained materials using order statistics and deep neural network architectures, massive ultrasonic data compression using wavelet packet transformation optimized by convolutional autoencoders, and software-defined ultrasonic system design for communications through solid structures.

This short course will provide an overview of the basic techniques of Doppler blood flow imaging used in the industry, followed by the limitations of these conventional approaches to imaging low-velocity blood flow. The course will then cover how low-velocity flow detection has been greatly improved with the advent of new microvascular Doppler flow imaging modes found on a number of commercial imaging systems. The second part of the course will cover how the introduction of microbubbles solves a key limitation of even these new microvascular Doppler modes to allow visualization of relative variations in microvascular flow. However, ultrasound contrast has had limited clinical application, in part due to a lack of quantification. Finally, traditional and new efforts to quantify microvascular hemodynamics using microbubbles, particularly nonlinear ULM approaches, will be presented.