Burlington, Massachusetts-based Butterfly Network develops handheld ultrasound technology. B-lines on an ultrasound scan appear as bright, vertical lines indicating wetness in the lung, associated with pulmonary air-space disease. That includes congestive heart failure, chronic obstructive pulmonary disease (COPD), pneumonia and COVID-19.
The auto B-line counter leverages deep learning technology to produce a B-line count from a six-second ultrasound clip. This differs significantly from the traditional manual counting processes to interpret B-lines. According to Butterfly Network, this could simplify how healthcare professionals evaluate adults with suspected diminished lung function. This could accelerate their ability to make informed treatment decisions at the point of care.
“Our goal at Butterfly is to give healthcare practitioners, and eventually consumers, a real-time full color, annotated, window into the human body. Applying AI to make ultrasound easier to use is core to Butterfly, and will enable powerful ultrasound to be in the palm of more clinician’s hands, across specialities, to monitor, assess, and prescribe treatments in a more informed way,” said Dr. Jonathan Rothberg, Butterfly Network founder and interim CEO. “Our AI-enabled Auto B-line Counter empowers providers to assess lung conditions faster and with more confidence – and in turn, will aid in earlier detection, diagnosis, and treatment of cardiovascular diseases, a leading cause of death globally, taking nearly 18 million lives each year.”
More about the Butterfly Network auto B-line counter algorithm
The auto B-line counter algorithm utilizes state-of-the-art instant percent counting. This assigns numbers to confluent B-lines by the percentage of rib space occupied in addition to counting discrete B-lines. This technique proves more reliable than incumbent individual line counting methods, Butterfly Network says.
Using this algorithm, trained providers can simply place the probe and receive a reliable number count directly on their screen.
Butterfly Network developed and trained its AI algorithms through the secure Butterfly Cloud. This offers access to more than 3.5 million de-identified ultrasound cines. These data inputs come from hundreds of sites across the U.S. and include a diverse range of age, gender, body mass, ethnicity and race.
The company anticipates a launch for the AI-enabled auto B-line counter in the U.S. in the “early summer.”
