A new type of ultrasound technology developed by AUSTRAL Diagnostics can detect small changes in the lungs that could be indicative of lung disease.
Mathieu Couade, chief scientific officer at AUSTRAL, a medtech startup based in Paris, and colleagues have built an airborne ultrasound surface motion camera that is able to detect small vibrations in the chest known as vocal fremitus.
Physicians are trained to listen to chest vibrations using stethoscopes to try and pick up signs of lung conditions such as pneumonia, which changes the sound of a person’s breathing.
“Physicians compare the sensory elements perceived during physical examination with a personal repertoire resulting from initial learning and accumulated practice. This allows them to identify abnormalities deriving from disease-related changes in the structural characteristics of the respiratory system,” write the authors.
“The value of physical examination of the thorax depends on the skills and experience of the operator but is generally low. Indeed, the performances of the human senses limit it, as does its fragmentary nature (localized sampling). In addition, the collected data cannot be recorded for transmission, external assessment, and use for future reference.”
Writing in the journal AIP Advances, Couade and team describe results from a study using their airborne ultrasound camera, which doesn’t require skin contact, in 77 healthy volunteers (both men and women). The airborne camera allows the observation of the human thorax surface vibrations due to respiratory and cardiac activities at high frame rates of typically 1,000 images per second, according to Couade.
The researchers successfully mapped the vibrations caused by the vocalizations of the volunteers using their camera. They plan to use the data to map “normal” lung sounds and vibrations before using the device to map vibrations linked to different lung and cardiovascular diseases.
“The spatial distribution of vibrational energy was found to be asymmetric to the benefit of the right size of the chest, and frequency dependent in the anteroposterior axis,” said Couade, in a press statement. “As expected, the frequency distribution of vocalization does not overlap between men and women, with the latter being higher.”
The researchers want to collect as much data on vibrations in healthy and unwell people as possible so they can use artificial intelligence to help diagnose specific lung conditions based on an ultrasound scan with their airborne ultrasound camera in the future.
“Massive data collection of vocal fremitus could allow using artificial intelligence algorithms to isolate vibration patterns that could help disease identification. Up to now, in contrast, the interpretation of vocal fremitus is subject to the physician’s experience and remains subjective,” conclude the authors.
