Researchers at the University of California, Los Angeles (UCLA) have just developed an intelligent electronic patch that could restore speech to people who have lost it temporarily or permanently. This non-invasive medical device, the size of a coin, is capable of recording throat movements and transforming them into speech using artificial intelligence. A promising advance to improve the quality of life of patients suffering from voice disorders.
A self-powered patch that senses throat movements
The patch developed by the UCLA team measures only 6.5 cm2 and weighs only 7 grams.
Placed at the level of the larynxit is equipped with micromagnets under its surface which create an electromagnetic field through the movements of the intrinsic muscles of the larynx. These movements are then transformed into detectable electrical signals.
The device is completely self-powered with electrical energy thanks to a process based on the magneto-elasticity of the material used. Its magnetic field varies depending on the deformation, a variation captured by a coil via theelectromagnetic induction.
This is enough to generate the electricity needed for motion detection, without the need for an additional battery.
Artificial intelligence translates signals into audible words
Once the muscular movements have been captured and transformed into electrical signals, it is the turn of theartificial intelligence to enter the stage. A specially trained algorithm will classify and analyze this voice signal, then broadcast it via a tiny loud speaker embedded in the patch.
To train the AI, the researchers conducted tests on eight participants without voice disorders. They had to place the patch on their throat and repeat short phrases like “Merry Christmas”, “I love you”, etc. 100 times.
The algorithm then based itself on muscle movements to associate them with these parts of sentences. Result: the program recognized the sentences spoken with a 95% accuracy.
A portable, non-invasive solution for voice disorders
According to Jun Chen, assistant professor of bioengineering at UCLA, this smart patch presents a option portable et non invasive to help patients communicate while awaiting treatment, as well as during a recovery period after treatment for voice disorders.
Unlike existing solutions such as portable electrolarynx or tracheoesophageal puncture surgical procedures, which can be inconvenient, invasive or uncomfortable, this new device offers a much more practical and comfortable alternative for patients.
Towards commercialization of the patch?
If the first results are very promising, researchers now plan to conduct trials on actual patients suffering from various voice disorders. They expect to see similar results in terms of AI speech recognition accuracy.
The objective is to ultimately be able quickly equip patients whose laryngeal muscles present a pathology, whether benign dysphonia linked to a cold, disorders resulting from pathologies of the vocal cords or even following surgery for laryngeal cancer.
Before possible marketing, the patch will still have to go through regulatory processes, in particular approval of the FDA (Food and Drug Administration) in the United States. But this innovation “unquestionably showed its potential to help patients suffering from voice disorders” rejoices Ziyuan Che, researcher in the department of bioengineering at UCLA and co-author of the study.
My name is Michael. I am a journalist for this magazine where I analyze and break down current events in order to provide you with articles that will allow you to have as nuanced and fair an understanding of global issues as possible. Passionate about creating bridges between stories and people, I joined this site to practice interactive journalism, where each article is an open dialogue with the reader.
2024-03-31 08:51:47