Conventional ultrasound employs piezoelectric crystals that convert electrical energy into ultrasonic waves – vibrations – and the transducer serves as both transmitter and receiver. These systems suffer from two drawbacks: first, separate probes are required for different clinical applications; and second, a conventional probe requires a multitude of wires to connect the piezoelectric transducer to its electronic control system. In the iQ, the piezoelectrics have been replaced by a capacitive micromachined ultrasound transducer (CMUT) that acts as a tiny drum to generate the vibrations.
Rothberg and his team have taken advantage of the mass-market fabrication techniques for computers to develop their “ultrasound on a chip” (seen at left). He says, “We put all the elements onto a semiconductor wafer, then we dice up the wafer to make 48 ultra low-cost ultrasound machines.” The company plans to sell the iQ – that has been cleared by the FDA for 13 clinical applications, including breast imaging – for just $2,000, which means its patented technology has the potential to disrupt the point-of-care market.
In addition to its portability and attractive price, the iQ’s other major selling point is its application of artificial intelligence (AI) for both image acquisition and analysis. The engineers at Butterfly have trained the iQ’s software on large datasets of ultrasound images so that the system knows the difference between high- and poor-quality images. In addition, the software is able to make simple measurements – such as the ejection fraction of the heart – that will assist a clinician’s diagnosis.
The company anticipates the iQ will be used in rural health clinics in the developing world, where staff may not have the necessary expertise with ultrasound, but in time it could also become a consumer product in the developed world. It is clear the Butterfly Network will have to sell many iQ systems at $2,000 to recoup its initial investment of $100 million. Nevertheless, 2017 may just be remembered as the year that “ultrasound on a chip” made its debut.