By Richard L. Gaw

Staff Writer

For many of us, determining our life’s aspirations is an often topsy-turvy roller coaster ride that eventually – whether by influences or by accident -- stumbles onto something. By the time he was in the fifth grade at Kemblesville Elementary School, Bill Lotter knew what he wanted to do.

In his essay, “What I Want to Do When I Grow Up,” Lotter wrote that he wanted to become the offensive coordinator for the Philadelphia Eagles, his favorite team. He loved sports, especially football, and in his essay, Lotter wrote that if he couldn’t work for the Eagles, he’d settle for becoming an ESPN announcer.

While these aspirations never came to pass, the early interest Lotter had developed in math and science did, and recently, they helped lead the way to a discovery that may eventually save the lives of many of the more than 270,000 American men and women who are diagnosed with breast cancer each year.

Lotter, 32, the chief technology officer for Boston-based DeepHealth -- an artificial intelligence (AI) software company he co-founded in 2016 that provides assistance in the interpretation of medical images – recently published the results of a three-year study that illustrates an ability to detect breast cancer a year or more earlier than current practice. The study was published in Nature Medicine.

In its findings, DeepHealth, now a subsidiary of RadNet -- the largest supplier of outpatient imaging services in the U.S. -- compared its AI to five full-time, breast-fellowship-trained expert radiologists reading the same screening mammograms.

The company’s software exhibited higher performance than all five radiologists, and the results suggest that the AI could help detect cancer one to two years earlier than standard interpretation in many cases.

“Our results point to the clinical utility of AI for mammography in facilitating earlier breast cancer detection, as well as an ability to develop AI with similar benefits for other medical imaging applications,” Lotter said. “By building AI software with high performance, DeepHealth has the potential to help radiologists enable more widespread access to high quality care.”

While Lotter said that he and his colleagues – who include co-founder Greg Sorensen, a radiologist and former Harvard professor – anticipate that their AI software package for 2-D and 3-D mammograms will indeed save many lives through earlier breast cancer detection, there are still many hurdles to ascend before it is formally introduced into the marketplace.

The first step will be receiving FDA approval, which Lotter said is expected to come in 2021 for their first product.

“Our results are encouraging and we see the potential for impact, but we need to fully bring our technology to fruition and deliver those outcomes clinically to really make a difference,” he said. “Clinical deployment will be a next main phase and we’re working hard to ensure we do so effectively, including having frequent conversations with radiologists about integration, and also scaling up our computational resources to deploy our software across many clinics.”

‘Math and science nerd’

While the results of DeepHealth’s discovery may soon revolutionize how mammography screenings are conducted in the future, the journey that got Lotter there was developed early in his life, when he was a young boy growing up in Landenberg. A self-described “math and science nerd,” Lotter flipped his time between his passion for sports and the multidisciplinary connection between math and science.

When he wasn’t on the gridiron playing for the Red Devils at Avon Grove High School, Lotter was attending AP classes in biology, chemistry, physics, calculus and statistics. Although he was influenced by many of his teachers, one stands out above the rest: Gary Habbart, his calculus teacher.

“I appreciated how much Mr. Habbart cared about his students and what he taught. It was clear he really thought through each lecture and how to present it. This helped me develop my interests as well and gave me a solid math foundation going into college,” Lotter said. “I learned early on that I like trying to understand how things work and thinking about math and science, partly because they are not subjective. The same equations apply no matter who you are or where you’re from.”

After graduating from Avon Grove in 2006, Lotter attended Northwestern University, where he joined the university’s Integrated Science Program, and graduated with a B.A. in Integrated Sciences, a B.A. in Math and an M.S. in Applied Math.

By the time he entered Harvard University in 2013 to pursue a Ph.D. in Biophysics (with a secondary degree in Computational Science), Lotter wanted to focus on how he could bring math and science together in order to make a positive impact on people’s lives. 

“It was a perfect fit for me because it allowed me to work on multidisciplinary scientific problems with an opportunity for tangible impact,” he said.

Lotter complimented his studies by nurturing his fascination for machine learning that opened the door to developing machine learning solutions. Before starting at Harvard, he worked for two years as an algorithmic trader for a Chicago-based trading firm, and while at Harvard, he wrote a few articles for Harvard’s Sports Analytics Club that focused on developing algorithmic models to predict NFL player performance. Soon after, NFL teams began noticing, and it helped him land a position as a machine learning consultant for an NFL team. As he was finishing his studies, several other teams wanted Lotter to work for them.

One of them was the Philadelphia Eagles. In 2017, they offered him a job. It would be the opportunity he had dreamed about since he was a kid. Lotter accepted the offer at first, but then changed his mind.  

“This was the same season the Eagles won the Super Bowl, and if I had taken the job, that would have been an unbelievable experience,” he said. “I had initially told those I interviewed with that I had just begun a start-up company back in Boston, and that I was reluctant to leave what Greg [Sorenson] and I had just begun. It was the hardest decision of my life, and as much as I love the Eagles, in the end I just couldn’t give up on the start-up.

“It was still very early on for DeepHealth, but it was hard for me to give up its potential. Now a few years later, I’m proud of our team for being in a position to positively impact patients’ lives.”

To contact Staff Writer Richard L. Gaw, email [email protected].