Entrepreneur Jonathan Rothberg says he is developing a scanner the size of a smartphone that you could just hold up to a person’s chest and see a moving, 3D ultrasound image of what’s inside.

Remember that massive MRI machine at the hospital that’s used to search for tumors in the kidneys? Or the expensive ultrasound machine to have a look at your unborn baby? All of them could be passé if Jonathan Rothberg has his way.

Butterfly Network, a three-year-old tech startup owned by Rothberg, has just announced financing to the tune of $100 million, and hopes to build its first prototype of a handheld medical imaging device the size of a smartphone in the next 18 months.

Unlike many tech-entrepreneurs who fail to push through that tricky part of converting ideas into commercially viable products, Jonathan Rothberg has a reputation. This chemical and biomedical engineer has a knack for using semiconductor technology to solve the problems of biology in an innovative way.

Rothberg’s previous startups, 454 Life Sciences and Ion Torrent Systems—which he sold later for more than $500 million—are credited with bringing down DNA sequencing costs to less than a thousand dollars. Now he intends to shrink a room-sized MRI machine to a smartphone-sized scanner that will cost only a few hundred dollars.

While Rothberg remains tightlipped about how he plans to do that, patent applications filed by Butterfly Network give a glimpse into the technology behind the phone-sized scanner.

A typical ultrasound works on the principle of echolocation—the same technique used by bats, whales and dolphins to move around. The machine emits high frequency sound waves which are reflected back with varying intensities by different tissue boundaries in the body. These echoes are collected and then processed to display the distances and intensities of the echoes on a screen, forming an ultrasound image.

Most ultrasound machines are made of small ceramics or piezoelectric crystals to generate sound waves—but fabrication and wiring of these crystals is complicated. These are attached to a separate box to process the signals which, of course, increases the size and costs of the machine.

If the ultrasound elements can be directly integrated onto a computer chip, then they can be mass-produced at cheaper cost and even the large arrays required for creating a 3-dimensional image, instead of the 2-dimensional ultrasound image, can also be easily created.

It appears that Butterfly Network is placing its bets on creating just such a device, called a ‘capacitative micro-machined ultrasound transducer,’ which lets ultrasound emitters be etched directly onto a semiconductor wafer, along with other circuits and processors.

While this idea has been around for some time, no one has been able to come up with a workable, market-validated product. Rothberg, it seems, is up for the challenge and more. He wants to combine the ultrasound-on-a-chip technology with cloud computing, telehealth and deep learning to potentially disrupt the way medical imaging is done now.

First, the imaging systems are expected to be mass-produced—so cheap that they would be affordable even in the poorest regions of the world. Secondly, the diagnosis part is automated using software and cloud computing services. The images taken by the scanner are uploaded to a cloud which will then recognize patterns to identify possible onset of disorders.

Using the same or similar cloud services, the medical imaging systems can also be deployed in remote locations and integrated using telemedicine systems. Rothberg believes that using the device would be as simple as taking ‘panorama’ picture on an iPhone, requiring minimal training for a technician to operate the device.

Thirdly, the system will make use of ‘deep learning’, an artificial intelligence technique which means that the more the system is used, the smarter it gets in identifying the underlying patterns in the ultrasound images, making diagnosis faster and better.

 “When I have thousands of these images, I think it will become better than a human in saying ‘Does this kid have Down syndrome, or a cleft lip?’ And when people are pressed for time it will be superhuman,” says Rothberg.  “It will make a technician able to do this work.” 

"Spending 18 percent of our GDP, or $10,000 per person, per year, on healthcare is unsustainable and out of touch with the needs of the rest of the world," adds Dr. Rothberg. "Our mission is to democratize healthcare by launching companies, building devices, and combining advances in biology, semiconductors, and artificial intelligence. We built Butterfly to transform the way we image the body and perform surgery."

Shiva Gopal Reddy has a Bachelor's degree in Physics and a Master's in Applied Psychology and writes frequently on the latest research, impact, happenings and trends in digital health technology.