AI-powered sensor performs heart tests on the go

A non-invasive device capable of performing on-the-go ultrasound scans of the heart, both of structure and function, is what engineers and doctors are developing at the University of California San Diego. The new heart monitor, roughly the size of a postage stamp, can be worn for up to 24 hours even during strenuous exercise.

According to Sheng Xu, a professor of nanoengineering who leads the project, the goal is to make ultrasound scans more widely accessible as a replacement for echocardiograms, which need trained technicians and bulkier equipment. “This technology allows anyone to capture ultrasound images anywhere. The increasing risk of heart disease requires more advanced and inclusive monitoring procedures. By giving patients and physicist more complete details, this continuous, real-time monitoring of cardiac images can optimize and fundamentally reshape the paradigm of diagnostics,” highlights Xu.

The adhesive sensor is as soft as skin and does not limit patients’ movement, and can be affixed to the chest with minimal restriction on exercise. It sends and receives ultrasound waves used to produce a constant stream of images of the heart structure in real time.

The solution uses Artificial Intelligence algorithms to measure how much blood the heart is pumping. Insufficient blood pumping is usually the cause of most cardiovascular diseases. By being able to perform ultrasound scans anywhere, the new device can help make more accurate diagnoses of heart function, as problems frequently only manifest themselves when the body is in motion.

“A deep learning model automatically segments the shape of the left ventricle from continuous image recording, extracting the volume frame by frame and producing waveforms to measure systolic volume, cardiac output and ejection fraction,” explains Mohan Li, a master’s student in Professor Xu’s group at the University of California San Diego.

The cardiac monitoring sensor continuously captures images of the four chambers of the heart at different angles and analyses a clinically relevant subset in real-time with the help of Artificial Intelligence. In comparison, current methods have restricted sampling capabilities and provide limited data. In addition, the new technology minimizes patient discomfort and avoids the radiation exposure required in scans such as CT and PET scans.

The project, which builds on the team’s previous advances in wearable imaging technologies for deep tissue, was featured in the January 25 issue of the journal Nature. In the current phase of the project, the sensor is connected via cables to a computer to transfer the data. The team has already developed a wireless circuit that will be covered in a forthcoming publication of the study.

The intention is to commercialize the technology through Softsonics, a spin-off from the University of California San Diego founded by Professor Xu together with engineer Shu Xiang.

Beyond the heart

“Smart patches” are also being researched to detect certain biomarkers in the skin and identify early neurodegenerative diseases such as Parkinson’s and Alzheimer’s. A scientist at Swansea University in the UK has developed a technology using microneedles designate to break the skin barrier in a minimally invasive way and monitor biomarkers with clinical importance.

“The skin is the largest organ in the human body and contains more fluids than blood. These fluids have biomarkers that complement those of other biofluids such as sweat, saliva, and urine. They can be sampled in a minimally invasive manner and used in tests or analysed in real time using microneedle patches,” Sanjiv Sharma, who developed what is considered the world’s first ‘smart patch’ for Covid-19.

The sensor has a short measurement time of 6 minutes, with high accuracy and a low detection limit. According to the researcher, this new diagnostic tool will make the skin a window into the body and vital organs such as the brain.

Swansea University is working with research partners in the UK, Portugal, France, and Japan to deepen the field of transdermal diagnostics and extend it to the development of diagnostic devices for various other healthcare applications.

The electronic skin patches market is expected to grow at a compound annual rate of 20.17% between 2022 and 2027. In addition to universities, start-ups are working to offer electronic skin sensors to measure heart and respiratory rate or even detect blood alcohol levels from sweat. These patches could help monitor patients remotely and collect data outside laboratory and hospital settings. Diabetes condition monitoring accounts for the largest share of smart patches (43.51%), and this market segment is estimated to reach $18.1 billion by 2027, globally speaking.