Medgaget Interview: 4Dx Uses Algorithms to Better Visualize Lung Function

Chronic obstructive pulmonary disease affects millions of people in the United States. The main method of diagnosis is the pulmonary function tests (PFTs), in which a patient breathes into a machine that measures pulmonary parameters. The disadvantage, however, is that pulmonary function tests take an “average” of a patient’s lung without being able to detect specific areas of lung function and compare them over time.

4Dx is hoping to improve that. Using principles of air flow dynamics and applying them to data from a simple X-ray, the company’s algorithms can calculate the amount of air that each area of the lung is receiving. This generates a moving, color-coded visual of a patient’s breathing lung.

4Dx CEO Andreas Fouras

The unique thing about this visual, says founder and CEO Andreas Fouras, is that it gives much more information than a traditional pulmonary function test. Physicians can easily see exactly which areas of the lungs are poorly ventilated, whether certain areas are over-compensating for poor function in other areas, and compare specific areas of lung function over time. Plus, says Fouras, the visual is an incredibly easy way for patients to understand their disease.

The technology works by using data from X-rays. The patient breathes 3-5 breaths in front of an X-ray machine, at various angles, and that data is run through an algorithm that converts the moving changes seen in the X-ray to air flow in the lungs. “We look at how the lung moves from each of the few angles,” explains Fouras, “and then we piece that together not to form a three-dimensional picture of the lung, but to form a three-dimensional picture of how the lung moves—or a 4D picture.” People think of X-rays as a method of directly identifying structures, says Fouras, but there’s actually a lot more information there.

The idea began when Fouras was pursuing his PhD at Monash University in Australia, where he helped develop imaging technology for analyzing air flow dynamics in wind tunnels. At the same time, he happened to talk with various physicians and got the idea to apply the same technology to lung imaging. Encouraged by the enthusiastic initial reactions of physicians, he began working towards developing a technology specifically for the lungs.

The research led to many grants and published research papers. “My academic career went ballistic,” he quips, as he went from new PhD to tenured professor in six years. But to successfully commercialize the technology, he had to go all in. He sold his house, moved his family to Los Angeles, and borrowed money to invest in his fledgling company. Since its official founding in 2012, 4Dx has grown to a team of thirty people.

The company is currently awaiting FDA approval, which will be decided sometime this year. In the meantime, the company is reaching out to hospitals and working with physicians to do investigational studies. “The vast majority of people we’ve spoken to have come on board,” says Fouras, a promising sign reminiscent of the enthusiasm his research generated when he was a student and professor in Australia.

The technology doesn’t have to be limited to the lungs, either. “Once you have a solid platform that has a good performance, the opportunity to spread it out across other applications is really significant,” says Fouras. “I don’t want it to be limited by my imagination.” The algorithms could be tailored for better visualizing the heart and other organs — the company even has a patent on using the technology for in vitrofertilization.

“But the thing that encourages me the most is the response I get when I talk to doctors,” says Fouras. “I see the look in their eyes, and they say something like ‘where have you been my whole career? I can’t wait to use this.’” And so above all of the excitement for the technology and its future, he says, “That’s really the most exciting thing for me.”

The original article “4Dx Uses Algorithms to Better Visualize Lung Function”, by CiCi Zhou is available at Medgaget.com.