Novel use of a lung ultrasound sensor for monitoring lung conditions

Tanya Khokhlova – tdk7@uw.edu
Adam Maxwell – amax38@uw.edu
Gilles Thomas – gthom@uw.edu
Jeff Thiel – jt43@uw.edu
Alex Peek – apeek@uw.edu
Bryan Cunitz – bwc@uw.edu
Michael Bailey – mbailey@uw.edu
Kyle Steinbock – kyles96@uw.edu
Layla Anderson – anderla@uw.edu
Ross Kessler – kesslerr@uw.edu
Adeyinka Adedipe- adeyinka@uw.edu
University of Washington
Seattle, WA, 98195

Popular version of paper ‘1aBAb12 – Novel use of a lung ultrasound sensor for detection of lung interstitial syndrome’

Presented Monday morning, November 29, 2021

181^st ASA Meeting

The need to continuously evaluate the amount of fluid in the lung is essential in patients suffering from a number of conditions, including viral pneumonia (including COVID-19) and heart failure, and patients on dialysis. Chest x-ray and CT are typically used for this purpose, but can not be done continuously due to the radiation dose, and have logistical limitations in some cases, for example when transporting unstable patients or patients with COVID-19 due to the risk of contagion. Lung ultrasound (LUS) is non-ionizing and safe, and has recently emerged as a useful triage and monitoring tool for quantification of lung water. Because lung is air-filled, it is reflective for ultrasound, and in LUS exams it is image artifacts that are being evaluated, rather than true lung images. The artifacts termed A-lines are periodic bright horizontal lines parallel to the lung surface representing multiple reflections of ultrasound pulse from the lung and indicating a normal aeration pattern. The artifacts termed B-lines are comet-like bright vertical regions originating at the lung surface and extending down. The number and distribution of B-lines are known to correlate with presence of fluid in the lung and the condition severity. However, visualization and quantification of B-lines requires training and is machine and operator dependent, whereas in select clinical scenarios continuous, automated hands-free monitoring of lung function is preferred, e.g. COVID19 infection.
In this study we were aiming to identify the detected ultrasound signal features that are associated with B-lines and to develop a miniature wearable non-imaging lung ultrasound sensor (LUSS). Individual adhesive LUSS elements could be attached to patients in specific anatomic locations similarly to EKG leads, and ultrasound signals would be collected and processed with automated algorithms continuously or on demand. First, we used an open platform ultrasound imaging system to perform standard 10-zone LUS in ten patients with confirmed pulmonary edema, and in five healthy volunteers. The ultrasound signal data corresponding to each image were collected for subsequent off-line Doppler, decorrelation and spectral analyses. The metrics we found to be associated with the B-line thickness and number were peaks of Doppler power at the pleural line and the ultrasound signal amplitude corresponding to a large depth.

Left: examples of lung ultrasound images containing A-lines and B-lines and the corresponding signals detected by the ultrasound imaging probe. Right: conceptual diagram of the use of LUSS for monitoring of lung condition and a prototype LUSS element. Adhesive LUSS elements are applied in 10 anatomic locations and automated signal processing software displays lung fluid score for each element on a 4-point scale: none (green), mild (yellow), moderate (orange) or severe (red).

Next, we built miniature LUSS elements powered by custom-built multiplexed transmit-receive circuit, and tested them in a benchtop lung model – polyurethane sponge containing variable volumes of water – side by side with LUS imaging probe previously used in patients. Wetting of the sponge produced B-lines on the ultrasound images, and the associated ultrasound signals were similar to those measured by LUSS elements. We hope to proceed with testing LUSS in human patients in the nearest future. This work was supported by NIH R01EB023910.

Extracting Human Skull Properties by Using Ultrasound and Artificial Intelligence

Churan He¹– churanh2@illinois.edu
Yun Jing² – jing.yun@psu.edu
Aiguo Han¹ – han51@illinois.edu

1. Department of Electrical and Computer Engineering
The University of Illinois at Urbana Champaign
306 North Wright Street
Urbana, IL 61801

2. Graduate Program in Acoustics
Pennsylvania State University
201 Applied Science Building
University Park, PA 16802

Popular version of paper ‘1aBAb9 – Human skull profile and speed of sound estimation using pulse-echo ultrasound signals with deep learning’

Presented Monday morning, November 29, 2021

181st Meeting of the Acoustical Society of America in Seattle, Washington.

Ultrasound is a tremendously valuable tool for medical imaging and therapy of the human body. When it comes to applications in the brain, however, the presence of the skull poses severe challenges to both imaging and therapy. The skulls of human adults induce significant distortions (also called phase aberrations) to the acoustic waves. The aberrations result in blurred brain images that are extremely challenging to interpret. The skull also distorts and shifts the acoustic focus, causing challenges in therapy of the brain (such as treating essential tremors and brain tumors) using high-intensity focused ultrasound.

Prior research has shown that phase aberrations can be most accurately corrected if the skull profile (i.e., thickness distribution) and speed of sound are known a priori. Various methods have been proposed to estimate the skull profile and speed of sound. The gold-standard method used in treatment planning derives the skull properties from computed-tomography (CT) images of the skull. The CT-based method, however, entails ionizing radiation, potentially causing harm to the patients.

We propose an ultrasound-based method to extract the skull properties. This method is safer because ultrasound does not cause ionizing radiation. We developed an artificial intelligence (AI) algorithm (specifically, a deep learning algorithm) that predicted the skull thickness and sound speed by using ultrasound echo signals reflected from the skull.

We tested the feasibility of our method through a simulation study (Figure 1). We performed acoustic simulations using realistic skull models built from CT scans of five ex vivo human skulls (see animation). The simulations generated a large number (=7891) of ultrasound signals from skull segments for which the thickness and sound speed were known. We used 80% of the data to train our AI algorithm and 20% for testing. We developed and tested two algorithm versions: One version took the original echo signal as the input and the other used a transformed signal (i.e., Fourier transform that displays the signal’s frequency spectrum).

Both versions of our AI algorithm achieved accurate results, while the version using the transformed signals appeared to be more accurate. Using the original signal as the input, we obtained a mean absolute error of 0.3 mm for skull thickness prediction and 31 m/s for sound speed prediction. When transformed signals were used, the error in thickness prediction was reduced to 0.2 mm (= 3% of the average skull thickness [6.3 mm]), and the error in sound speed prediction was reduced to 25 m/s (= 1% of the average sound speed [2340 m/s]). In the case of transformed signals, the correlation between predicted values and the ground truth was 0.98 for thickness and 0.81 for speed of sound (Figure 2), where a correlation value of 1 represents perfect correlation.

Collectively, our preliminary results demonstrate that the developed AI algorithm can accurately estimate skull thickness and speed of sound, providing a potentially powerful tool to correct skull phase aberration for transcranial ultrasound brain imaging and therapy.

[Animation: 3-dimensional density map of one of the skulls used in the study]

Figure 1. Schematic diagram of the simulation study

Figure 2. a) Scatter plot of extracted speed of sound versus ground truth; b) scatter plot of extracted thickness versus ground truth.