A Fully Integrated Autonomous Wearable Ultrasonic System-on-Patch (USoP)

Andrew Corselli

A team at the University of California San Diego has developed the first fully integrated wearable ultrasound system for deep-tissue monitoring — including for subjects on the go. The technology facilitates potentially life-saving cardiovascular monitoring and marks a major breakthrough for one of the world’s leading wearable ultrasound labs.

“This project gives a complete solution to wearable ultrasound technology — not only the wearable sensor, but also the control electronics are made in wearable form factors,” said first author Muyang Lin. “We made a truly wearable device that can sense deep-tissue vital signs wirelessly.”

The fully integrated autonomous wearable ultrasonic system-on-patch (USoP) builds on the lab’s previous work in soft ultrasonic sensor design. However, prior soft ultrasonic sensors all required tethering cables for data and power transmission. However, in this work  , published in Nature Biotechnology, the USoP includes a small, flexible control circuit that communicates with an ultrasound transducer array to collect and transmit data wirelessly. A machine-learning component helps interpret the data and track subjects in motion.

The lab notes that the USoP allows continuous tracking of physiological signals from tissues as deep as 164 mm, continuously measuring central blood pressure, heart rate, cardiac output, and other physiological signals for up to 12 hours at a time.

“This technology has lots of potential to save and improve lives,” Lin said. “The sensor can evaluate cardiovascular function in motion. Abnormal values of blood pressure and cardiac output, at rest or during exercise, are hallmarks of heart failure. For healthy populations, our device can measure cardiovascular responses to exercise in real time and thus provide insights into the actual workout intensity exerted by each person, which can guide the formulation of personalized training plans.”

While developing its latest innovation, the team was surprised to discover that it had more capabilities than initially anticipated.

“At the very beginning of this project, we aimed to build a wireless blood pressure sensor,” said Lin. “Later on, as we were making the circuit, designing the algorithm, and collecting clinical insights, we figured that this system could measure many more critical physiological parameters than blood pressure, such as cardiac output, arterial stiffness, expiratory volume and more, all of which are essential parameters for daily health care or in-hospital monitoring.”

Moving forward, the sensor will be tested among larger populations.

“We are now in the process of working with clinicians to obtain IRBs (approval for clinical trial) from the university,” Lin said in an exclusive Tech Briefs interview, the entirety of which can be read below.

Here is the Tech Briefs interview with Lin and co-first author Ziyang Zhang, edited for clarity and length.

Tech Briefs: What inspired your research?

Lin: Although we are research engineers, we have a very close relationship with clinical collaborators. They always give us valuable feedback, and we do know the medical problems they face. In current clinical practice, ultrasound machines are bulky and wired. They require an experienced sonographer to perform manual probe maneuvering and require the subjects to remain motionless. This creates a challenge for the accessibility and accuracy of medical ultrasonography. Our new wearable ultrasound technology is a unique solution to address these challenges and enable long-term deep tissue monitoring on-the-go.

Tech Briefs: What was the biggest technical challenge you faced?

Zhang: The biggest challenge we encountered was the generalizability issue for our decision-making and diagnostic machine-learning algorithm. This is a well-known challenge for AI and medical image processing. When we train the machine-learning model on one subject’s data, the algorithm may not work on a new subject.

The difference in subject-dependent data features is unpredictable and sometimes it could significantly compromise the algorithm’s performance. We eventually made the machine-learning model generalization work by applying an advanced adaptation algorithm. This algorithm can automatically minimize the domain distribution discrepancies between different subjects, which means the machine intelligence can be transferred from subject to subject. We can train the algorithm on one subject and apply it to many other new subjects (with minimal retraining).

Tech Briefs: Can you explain in simple terms how the technology works?

Muyang Lin: We have made three major technological breakthroughs: First, we designed a wearable ultrasonic probe that can collect deep tissue signals from the skin surface. Second, we designed soft circuitry, which could automatically collect and transmit data wirelessly. Third, we made a decision-making, machine-learning algorithm for data analysis. With these advances, deep tissue physiology can be monitored in motion, which provides unprecedented opportunities for medical ultrasonography and exercise physiology.

Tech Briefs: The paper I read says, ‘Moving forward, the sensor will be tested among larger populations.’ Any research/work done on the clinical trials? How is that coming along?

Lin: We are now in the process of working with clinicians to obtain IRBs (approval for clinical trial) from the university.

Tech Briefs: The paper also says, ‘Xu is the co-founder of Softsonics, LLC, which plans to commercialize the technology.’ How soon do you think we will see your technology commercialized?

Lin: The commercialization and clinical trial are related. This is the next generation of health-monitoring devices, so validating on a larger population is necessary. Passing the clinical trial is the key. We anticipate that it may take one year to validate our sensor against existing gold-standard medical devices. After we receive the clearance from clinical trials, this sensor could be commercialized and applied in clinics.

Tech Briefs: Do you have any advice for engineers aiming to bring their ideas to fruition?

Lin: To bring a rough idea into pioneering research, there are three key steps in general — know the problem, understand the technology, and keep innovating.

Although we are engineers, we do know the medical problems that clinicians face. We have a close relationship with clinicians and learn insights from them. Solving practical challenges brings high impact on the research.

A good understanding of existing technologies is also critical. There are technical barriers to be overcome, but to identify these barriers we need to understand the challenges. For example, one technical barrier for wearable ultrasound is that existing probes are operator dependent. Our latest development overcomes this challenge by implementing an autonomous algorithm to mimic clinicians’ knowledge to make decisions and diagnoses.

Innovation is also the key to developing revolutionary technology. We always keep our minds open to new technologies. Ideas from other fields are always inspiring. In this project, we borrowed object recognition algorithms developed by computer scientists to help us process medical images.

Tech Briefs: Anything else you’d like to add?

Lin: Existing wearable sensors are only focusing on the signals on the skin surface or shallow under the skin. We are going to deep tissues, whose signals have a stronger and faster correlation to those shallow signals. UCSD is pioneering this research direction and there are quite a few research groups in the world starting to do similar work now.

---