Scott Schoen Jr – scottschoenjr@gatech.edu
Costas Arvanitis – costas.arvanitis@gatech.edu

Georgia Tech
901 Atlantic Dr
Atlanta, GA 30318

Popular version of 2pBAc – Spatial Characterization of High Intensity Focused Ultrasound Fields in the Brain
Presented Tuesday afternoon, December 8, 2020
179^th ASA Meeting, Acoustics Virtually Everywhere

The pitch and size of a sound are quite intrinsically connected. This is why, for instance, low-register instruments (such as a tuba or double bass) are large, while higher pitched ones may be very small (like a piccolo or triangle). Sound travels in waves, and the product of the length of the wave (wavelength) and its pitch (frequency) is a constant (namely, the speed of sound).

Consequently, the wavelength of sounds we can hear may be between about 15 m and 0.2 cm. But just as there are wavelengths of light we cannot see (such as ultraviolet and X-rays), there exists sound with much smaller wavelengths. Ultrasound, so called since its frequency is above our hearing range, is able to travel through human tissue and enables noninvasive imaging with millimeter resolution.

Since sound is pressure, it also carries energy. And, much like sunlight through a magnifying glass, sound energy may be focused to a small area to cause heating. This technique has allowed noninvasive and minimally invasive therapy, where focused ultrasound (FUS) creates small regions of high heat or forces to burn or manipulate the tissue. This is especially important for brain diseases, where surgery is particularly challenging.

Fig. 1 – Human cells are sensitive to sound frequencies from about 20 Hz to 20 kHz (left). However, focusing sound to a small area requires small wavelengths—and thus much higher frequencies (right). Not to Scale

Interestingly, it turns out that at very high pressures, so-called nonlinear acoustic effects become important, and the sound begins to interact with itself. One consequence is that if the FUS has to very high frequencies, say 995 kHz and 1005 kHz, the focal spot will a few millimeters, similar to 1 mm. However, the high pressure interaction will also generate energy at 1005 kHz – 995 kHz = 10 kHz—within the audible and tactile range.

This work describes our use of simulations and experiments to understand how this low frequency energy might be realized for FUS through the skull. Understanding the strength and distribution of low frequency energy generated with high frequency FUS may open a new range of therapeutic and diagnostic capabilities in one of the most complex and medically imperative organs: the brain.

Tuhin Roy – troy@ncsu.edu

Murthy Guddati – mnguddat@ncsu.edu
NC State University – Civil Engineering, Raleigh, NC 27695, USA

Matthew W. Urban – Urban.Matthew@mayo.edu

James Greenleaf – jfg@mayo.edu
Mayo Clinic – Department of Radiology, Rochester, MN 55905, USA

Popular version of paper 1aBAd1:  Guided wave inversion for arterial stiffness
Presented Monday morning, December 7, 2020
179th ASA Meeting, Acoustics Virtually Everywhere

Cardiovascular disease is a leading cause of death in the United States and worldwide. Atherosclerosis, or the stiffening of arteries, contributes to damage of downstream organs such as the brain and heart. If early atherosclerosis can be identified, it may be treated. Our research is motivated by developing a diagnostic tool for early detection of atherosclerosis using one of the cheapest and safest modalities, medical ultrasound, which can used widely across the world.

The target for this work is on estimating the stiffness and other mechanical properties of the carotid artery, a well-known indicator of cardiovascular disease. To accomplish this aim, we use a technique called shear wave elastography, where the wave propagation characteristics measured in the arterial wall are used to estimate the stiffness of the artery. Specifically, we use acoustic radiation force, resulting from focused ultrasound waves from an ultrasound probe to tap on the wall of the artery. This tap creates waves that travel within the artery wall, which are also measured with the same ultrasound probe. In this work, we present algorithms that convert the wave motion measured with ultrasound to values of arterial stiffness.

Christopher Pacia – cpacia@wustl.edu
Lifei Zhu
Jinyun Yuan
Yimei Yue
Hong Chen – hongchen@wustl.edu

Washington University in St. Louis
4511 Forest Park Ave
St. Louis, MO 63108

Popular version of paper 3pBAb1
Presented Wednesday afternoon, December 9, 2020
179^th ASA Meeting, Acoustics Virtually Everywhere

Brain cancer diagnosis starts with magnetic resonance imaging, or MRI, which allows clinicians to locate a tumor in the patient’s brain. However, MRI only provides anatomic information about the brain tumor. To understand the tumor type and to make a decision about future treatment, a neurosurgeon performs a tissue biopsy, drilling a small hole in the skull and carefully extracting a tumor sample with a long hollow needle. Liquid biopsy uses a blood sample to achieve similar information as the brain biopsy, without the need for surgery.

Unlike other cancers, whose small biomarkers, such as DNA, can be found circulating in a patient’s blood, brain cancers are separated from the rest of the body by the blood-brain barrier that does not allow tumor DNA to seep into the blood circulation. Two technologies are combined to briefly open the barrier: focused ultrasound and microbubbles. Focused ultrasound uses low-frequency ultrasonic energy to target tumors deep in the brain. Microbubbles are tiny gas bubbles commonly used in ultrasound imaging. When microbubbles are injected into a blood vessel, they travel along the blood flow to all parts of the patient’s body, including the brain. Once at the brain tumor, focused ultrasound causes the bubbles to expand and contract against the blood vessels in the brain, disrupting the blood-brain barrier and opening a door for the tumor DNA to be released into the blood circulation.

Video demonstrating the sonobiopsy technique to diagnose brain cancer.

The research presented here proves the success of sonobiopsy in increasing the levels of brain tumor biomarkers in the blood for the diagnosis of the most common and deadly brain tumor, glioblastoma, with different biomarker types and animal models. Sonobiopsy was optimized by increasing the amount of ultrasonic energy and the number of microbubbles injected to improve the number of biomarkers released in a mouse model. The utility of sonobiopsy was extended to different sized tumors and may be more effective for larger tumors, as demonstrated in a rat model. The potential for clinical translation was demonstrated by enhancing the release of brain-specific biomarkers in a pig model, with similar skull thickness as humans.

Sonobiopsy may be integrated into future clinical practice as a complement to MRI and tissue biopsies as an approach to noninvasively acquire molecular information of the tumor. The potential impact can be for the diagnosis of not only brain tumors but all other brain diseases. There are more studies to be done to better understand and optimize the technology before its practical value in humans, but this presentation is a step towards the future of brain cancer diagnosis.

Timofey Krit – timofey@acs366.phys.msu.ru
Arina Ivanova – ivanova.ad16@physics.msu.ru
M.V. Lomonosov Moscow State University, Faculty of Physics
Vorobyovy Gory, 1/2, Moscow 119991, Russia

Yuly Kamalov – kamalov53@yandex.ru
Russian Scientific Center of Surgery named after academician B.V. Petrovsky
Abrikosovsky Lane, 2, Moscow 119991, Russia

Popular version of paper 1aBAb
Presented Monday morning, December 7, 2020
179th ASA Meeting, Acoustics Virtually Everywhere

Ultrasound is a widespread diagnostic method due to its ease of use and relatively low cost of equipment. However, it is impossible to accurately determine the state of viscera using ultrasound. That is because the method is based on the detection of the boundaries between media with significantly different elastic properties. But ultrasound allows to track the movement of the medium precisely. More than two decades ago, academician of the Russian Academy of Science, professor O.V. Rudenko proposed a method in which a shear wave is excited at a certain depth. The shear wave velocity is defined primarily by the shear modulus of the medium. This elastic parameter changes significantly when the functional state of the medium changes. The values of the shear modulus of different organs range from several pascals to several gigapascals (Fig. 1). Therefore, the registration of the propagation of a shear wave, in contrast to ultrasound, allows one to determine the functional state of body organs and tissues.

Figure 1: The values of the shear modulus of different body organs

Skeletal muscle is a rather complex and unique entity. Many widely used elastic models are inapplicable to them. In this work, we used the above-mentioned method for measuring the shear modulus of skeletal muscles. The method was modified by focusing the ultrasonic beam into a shape called “blade” (Fig. 2).

Figure 2: The ultrasonic beam focused in a shape of “blade”

When focusing the ultrasonic beam, the “blade” shape makes it possible to excite a shear wave along the ultrasonic probe only. The excited wave then propagates to the left and to the right from the axis of symmetry of the ultrasound probe (Fig. 2). In anisotropic media, including skeletal muscles, this approach increases the measurement accuracy. “Blade” shape in the beam focus is not currently used in existing clinical equipment. However, measurements of the shear wave velocity in skeletal muscles can still be carried out on existing equipment that use the SWEI algorithm built into the ultrasound diagnostic device. For this purpose, the ultrasound probe is first placed along and then across the muscle fibers. Muscle fibers are visible in the standard B-mode, which allows measurements at two specified probe positions.

We obtained experimental data in healthy volunteers, whose biceps were loaded with the barbell plates. And with several loads, we measured the shear modulus along and across the muscle fibers of the biceps (Fig. 3).

Figure 3: The clinical trials in healthy volunteers

It turned out that with an increase in the load on the biceps, the shear modulus along the muscle fibers increases nonlinearly. The shear modulus across the muscle fibers does not depend on the applied load. The algorithm is clinically tested. It can be used to determine the functional state of skeletal muscles. The proposed method can already be used today to assess how various physical activities and nutrition systems affect skeletal muscles, as well as to identify the location of shear modulus inhomogeneities that cause muscle malfunction.

Muyinatu Bell — mledijubell@jhu.edu

Johns Hopkins University
3400 N. Charles St.
Baltimore, MD 21218

Popular version of paper 1aBAa4

Presented Monday morning, December 7, 2020

179th ASA Meeting, Acoustics Virtually Everywhere

Injuries to major blood vessels and nerves during surgical procedures such as neurosurgery, spinal fusion surgery, hysterectomies, and biopsies can lead to severe complications for the patient, like paralysis, or even death. Adding to the difficulty for surgeons is that in many cases, these bodily structures are not visible from their immediate viewpoint.

Photoacoustic imaging is a technique that has great potential to aid surgeons by utilizing acoustic responses from light transmission to make images of blood vessels and nerves. However, confusing artifacts that appear in the photoacoustic images, which are caused by acoustic reflections from bone and other highly reflective structures, challenge this technique and lead to inaccuracies in assumptions needed to form images.

Demonstration of ideal image formation (also known as beamforming) vs. beamforming that yields artifacts, distortions, incorrect localization, and acoustic reflections.

This paper summarizes novel methods developed by the Photoacoustic and Ultrasonic Systems Engineering (PULSE) Lab at Johns Hopkins University to eliminate surgical complications by creating more informative images for surgical guidance.

The overall goal of the proposed approach is to learn the unique shape-to-depth relationship of data from point-like photoacoustic sources – such as needle and catheter tips or the tips of surgical tools – in order to provide a deep learning-based image formation replacement that can more clearly guide surgeons. Accurately determining the proximity of these point-like tips to anatomical landmarks that appear in photoacoustic images — like major blood vessels and nerves—is a critical feature of the entire photoacoustic technology for surgical guidance. Convolutional neural networks (CNNs) – a class of deep neural networks, most commonly applied to analyzing visual imagery – were trained, tested, and implemented to achieve the end goal of producing clear and interpretable photoacoustic images.

After training networks using photoacoustic computer simulations, CNNs that achieved greater than 90% source classification accuracy were transferred to real photoacoustic data. These networks were trained to output the locations of both sources and artifacts, as well as classifications of the detected wavefronts. These outputs were then displayed in an image format called CNN-based images, which show both detected point source locations — such as a needle or catheter tip— and its location error, as illustrated below.

The well-trained CNN (top) inputs recorded sensor data from three experimental point-like sources (bottom left). This data produces an artifact appearing as a fourth source in with traditional beamforming (bottom middle). The CNN based image (bottom right) provides more clarity of true source locations and eliminates artifacts.

Overall, the classification rates ranged from 92-99.62% for simulated data. The network that utilized Resnet101 experienced both the greatest classification performances (99.62%) and the lowest misclassification rate (0.28%). A similar result was achieved with experimental water bath, phantom, ex vivo, and in vivo tissue data when using the Faster R-CNN architecture with the plain VGG16 convolutional neural network.

This success demonstrates two major breakthroughs for the field of deep learning applied to photoacoustic image formation. First, computer simulations of acoustic wave propagation can be used to successfully train deep neural networks, meaning that extensive experiments are not necessary to generate the thousands of example data needed to train CNNs for the proposed task. Second, these networks transfer well to real experimental data that were not included during training, meaning that the CNN based image can potentially be incorporated into future products that will use the photoacoustic process to minimize errors during surgeries and interventions.

Benjamin Wood – wood.benjamin@mayo.edu
Matthew W. Urban – urban.matthew@mayo.edu
Mayo Clinic Department of Radiology
200 First St SW
Rochester, MN 55905

Popular version of paper 3aBA6
Presented Wednesday morning, December 4th, 2019
178^th ASA Meeting, San Diego, CA

Introduction
Kidney stones affect approximately 12% of the global population as of 2018. Currently, the gold standard method of kidney stone location is computed tomography (CT) as the stones are easily visible because they have a higher Hounsfield unit due to the stone’s dense structure.

Currently, there are no other comparable imaging methods for noninvasively locating kidney stones. CT is limited in its use during kidney stone treatment as it is used sparingly in the initial location of stones and in post treatment to confirm if stones are still present. If stones are found early enough and have the correct composition, they can be treated with simple lifestyle changes like increased water intake and diet restrictions. Most often when symptoms of kidney stones arise, the stones are large enough that they are treated with surgical removal or lithotripsy.

Traditional B-mode ultrasound has historically been insufficient in locating kidney stones as it can be very difficult to distinguish stones from the surrounding tissue. Detection rates for ultrasound have been reported to be much lower than CT. In 1996, an artifact was discovered when using Doppler ultrasound that appears as a sparkling mosaic over the stone that was termed the twinkling artifact (TA). In recent years kidney stones have been tested as a clinical source of TAs. The goal of this present work was to explore how stone size and composition affect TAs and the ability to locate stones with TAs in an excised kidney.

Experiments

Isolated Stone Study
Initial experiments were performed using a wide range of stone types and sizes from 1.31-55.76 mm² in a cylindrical water tank with degassed water. Degassed water was used to reduce any introduction of microbubbles on the surface of the stones other than possibly due to ultrasound. Stones were suspended on a gauze bridge to limit TA appearance to the stones. All stones tested showed adequate TA signals regardless of stone composition or size.

Excised Kidney Study
To further test TA appearance on stones, they were individually place in an excised pig kidney and scanned in a large water tank with the same ultrasound probe as shown in Fig. 1. The power of the ultrasound pulses was tested to evaluate the ability to use the maximum power for initial location of the stones before lowering the power to a level that would precisely locate the stone and provide general information on its size. This showed no issues with the initial location of the stones with the TA.

Figure 1: Experimental setup for kidney stone scanning in an excised kidney.

Randomized Placement Study
A total of 47 stones were randomly placed within an excised kidney in a large water bath in groups of 5-8 stones per scan. This setup was used to evaluate the robustness of the method in a more clinical situation. The length of the kidney was scanned to locate as many stones as possible with some stones being placed next to each other purposefully. The process of locating and precisely pinpointing the stone is shown in Fig. 2. All 47 stones were located, including the stones placed in the same plane, with only two false positives.

Figure 2: Real-time Doppler scans of the TA over a calcium oxalate monohydrate that is 14.73 mm2 in cross-sectional area. The max voltage of 50 V was used for initial location and the minimum of 23.4 V was used for precision location.