People with amyotrophic lateral sclerosis (ALS) experience a gradual decline in their ability to control muscles. As a result, they often lose the ability to speak, making it difficult to communicate with others.

A team of researchers at MIT has now designed a stretchable, skin-like device that can be attached to a patient's face and measure small movements such as twitching or smiling. Using this approach, patients can convey various messages, such as "I love you" or "I am hungry," through subtle movements measured and interpreted by the device.

The researchers hope their new device will enable patients to communicate in a more natural way without having to deal with bulky equipment. The wearable sensor is thin and can be disguised as cosmetics to match any skin tone, making it unobtrusive.

LG Assistant Professor of Electronic Career Development, Department of Media Arts and Sciences, Massachusetts Institute of Technology, and head of the research team Canan Dagdeviren said: "Our devices are not only stretchable, soft, disposable, and lightweight, but also visually invisible." "You can camouflage it, and no one will think there is something on your skin."

The researchers tested the initial version of the device in two ALS patients (one female and one male, to ensure gender balance) and demonstrated that the device could accurately distinguish between three different facial expressions: smiling, opening the mouth, and pursing the lips.

MIT graduate student Farita Tasnim and former research scientist Tao Sun are the lead authors of the study, published today in Nature Biomedical Engineering. Other MIT authors are undergraduate Rachel McIntosh, postdoc Dana Solav, and research scientist Lin Zhang. Yuanyuan Gu and Nikta Amiri from the A * STAR Institute of Microelectronics in Singapore, Mostafa Tavakkoli Anbarani, and M. Amin Karami from the University at Buffalo are also authors.

Skin-like sensors

Dagdeviren's laboratory, the Conformal Decoder Group, specializes in developing conformal (flexible and stretchable) electronics that adhere to the body for various medical applications. After meeting Stephen Hawking (Stephen Hawking) in 2016, she became interested in researching how to help patients with neuromuscular diseases communicate. At the time, the world-renowned physicist visited Harvard University, where Dagdeviren was a junior fellow in the Harvard Society of Fellows.

Hawking passed away in 2018 from a slow-progressing form of ALS. He was able to communicate using an infrared sensor that detected cheek twitches, allowing a cursor to move across rows and columns of letters. While effective, this process could be time-consuming and required bulky equipment.

Other ALS patients use similar devices to measure the electrical activity of nerves controlling facial muscles. However, this approach also requires bulky equipment and is not always accurate.

"These devices are very rigid, flat, and square, and reliability is a major issue," said Dagdeviren. "Even with the same patient on the same day, you may not get consistent results."

Most ALS patients eventually lose the ability to control their limbs as well, so typing is not a viable strategy to help them communicate. The MIT team set out to design a wearable interface that allows patients to communicate more naturally without the bulky equipment required by current technologies.

The device they created consists of four piezoelectric sensors embedded in a silicone film. Made of aluminum nitride, the sensors detect mechanical deformation of the skin and convert it into easily measurable voltage. All these components are easy to mass-produce, so the researchers estimate the cost per device to be around 10 US dollars.

The researchers used a process called digital image correlation on healthy volunteers to help them select the most useful locations for placing the sensors. They randomly drew black-and-white spot patterns on the faces and then used multiple cameras to capture numerous images of the area while the subjects performed facial actions, such as smiling, twitching their cheeks, or forming shapes with their mouths for certain letters. The images were processed by software that analyzed how the small dots moved relative to each other to determine the strain levels experienced in individual areas.

"We had subjects perform different movements and created strain maps of various parts of the face," said McIntosh. "Then we looked at the strain maps, identified where on the face the appropriate strain levels for the device were observed, and determined that location to be suitable for placing the device for testing."

The researchers also used measurements of skin deformation to train a machine learning algorithm to distinguish between smiling, opening the mouth, and pursing the lips. They used this algorithm to test the device on two ALS patients and achieved approximately 75% accuracy in distinguishing these different actions. The accuracy rate for healthy subjects was 87%.

Enhancing communication

The researchers stated that based on these detectable facial movements, a library of phrases or words could be created to correspond to different combinations of movements.

"We can create customizable messages based on your movements," said Dagdeviren. "Technically, you can create thousands of messages, which is not possible with other existing technologies. It entirely depends on your library configuration, which can be designed for a specific patient or a group of patients."

Information from the sensors is sent to a handheld processing unit, which analyzes it using algorithms trained by the researchers to distinguish facial movements. In the current prototype, the unit is connected to the sensors, but the connection could also be implemented wirelessly, according to the researchers.

The researchers have filed a patent for this technology and now plan to test it on other patients. They stated that, in addition to helping patients communicate, the device could be used to track the progression of the disease or assess the effectiveness of treatments received.

"There are many clinical trials testing whether a specific therapy is effective in reversing ALS," said Tasnim. "The device can provide quantitative measures to track efficacy, rather than relying solely on patients reporting that they feel better or stronger."

This research was funded by the MIT Media Lab Consortium, the U.S. National Science Foundation, and the U.S. National Institute of Biomedical Imaging and Bioengineering.