Promising new AI can detect early signs of lung cancer that doctors can’t see

There are more than 300 FDA-approved AI tools for use in radiology, according to Anant Madabhushi, a professor in the department of biomedical engineering at Emory University School of Medicine in Atlanta. Most are used to help doctors diagnose and treat cancer, he said, but not to predict a person’s future cancer risk., like Sybil does.

Sybil looks for signs of where the cancer is likely to appear, so doctors know where to look and then can spot it as soon as possible.

Lung cancer is more treatable when caught early, said Dr. Kim Sandler, associate professor of radiology at Vanderbilt University Medical Center in Nashville, Tennessee.

But early detection is difficult, she says. Since the lungs cannot be seen or felt, the only way to spot them early is with a CT scan. By the time symptoms appear, including a persistent cough or difficulty breathing, the cancer is usually advanced and the most difficult to treat.

Pass research showed that screening with low-dose CT scans can reduce the risk of death from lung cancer by 24% because they can help detect cancer earlier, when it is easier to treat.

But an AI tool could potentially increase early detection rates for lung cancer — and potentially increase survival rates as well, Sandler said.

“It’s great for radiologists, but more importantly, it’s really great for our patients to be able to give them this tool to tell them whether or not we think it’s lung cancer,” she said.

How does Sybil work?

To predict cancer risk, Sybil relies on a single scan. It analyzes the three-dimensional image, looking not only for signs of abnormal growth in the lungs, but also for other patterns or nuisances that scientists don’t yet fully understand, said Dr. Florian Fintelmann, a radiologist at the Mass General Cancer Center and one of the researchers working on Sybil.

Based on what it sees, Sybil provides predictions for whether a person will develop lung cancer in the next one to six years, he said.

There have been cases, Fintelmann added, where Sybil picked up signs of cancers that radiologists didn’t pick up until nodules showed up on a CT scan years later.

Fintelmann said he sees a future in which the AI ​​tool helps radiologists make important treatment decisions, without completely replacing radiologists.

“The future of radiology will be assisted by AI,” he said. “You will always need a radiologist to identify where the cancer is, identify the best possible treatment, and actually do the treatment.”

The AI, however, is still far from perfect.

One issue that worries Madabhushi of Emory University is the type of data used to train the AI.

“A lot of data from medical institutions or clinical trials doesn’t represent the diversity of our country,” he said, adding that he believes AI tools are not being developed in a way tailored to help black and brown people.

The scientists who developed Sybil have recognized that the data used to create the AI ​​tool did not include “enough black or Hispanic patients to still be confident of broad applicability.”

The FDA has already taken action to address this issue, announcing last year that it would soon require researchers and companies seeking medical product approval to submit a plan to ensure diversity in clinical trials.

“We need to make sure that the AI ​​doesn’t reflect or capture our biases,” Madabhushi said.

There are also concerns about overdiagnosis, Sandler said. Doctors may give patients a potentially unnecessary biopsy for a nodule that may end up being benign.

“Do we find a cancer that we may not need to find? she says.

Follow BNC HEALTH on Twitter & Facebook.

Leave a Comment