More than 1.5 million new lung nodules are detected every year in the U.S., according to the Journal of Cardiothoracic and Vascular Anesthesia. While most are benign, determining which nodules require closer monitoring or biopsy is one of the most important steps in detecting lung cancer early, when treatment is most effective.
For years, that decision has been challenging. Lung nodules can be difficult to biopsy because the lungs are constantly moving, making it complicated to precisely target small lesions.
“When you're that deep into the lungs a single cough or movement can move you anterior, posterior, medial or lateral,” explained Huzaifah Salat, MD, a physician executive and pulmonologist in Milwaukee, Wisconsin, at Advocate Health.
Advocate Health is part of the AMA Health System Member Program, which provides enterprise solutions to equip leadership, physicians and care teams with resources to help drive the future of medicine.
Over the last decade, however, the advent of robotic-assisted bronchoscopy technology, cone-beam CT, augmented fluoroscopy and other advanced technologies have transformed physicians’ ability to evaluate even very small lung nodules with greater precision and lower risk.
“Today, I can reach an 8 mm nodule through a robotic bronchoscope that 10 years ago would have required surgery or a high-risk transthoracic needle biopsy—and then tell a patient that same day possibly what we're dealing with,” said Dr. Salat. “That is meaningful progress.”
In an interview with the AMA, Dr. Salat discussed how the profile of lung cancer is changing, the technological advances reshaping lung nodule evaluation and how to effectively use augmented intelligence (AI) technologies to enhance clinical expertise.
AMA: How has the patient profile of lung cancer changed through the years?
Dr. Salat: The profile of who gets lung cancer is shifting. We’re diagnosing lung cancer in younger patients, women and never-smokers at rates that challenge the traditional risk model. When a patient tells me they never smoked, I don’t treat that as reassurance—I take their symptoms seriously.
The gap between public perception and clinical reality is one of the most important things students entering this field need to understand.
AMA: How do you approach conversations with your patients who learn they have a lung nodule?
Dr. Salat: I try to explain to them what a lung nodule means. I counsel patients with empathy, while also being honest about the reality. I often start by telling them that the majority of small, sub-centimeter nodules are not cancer.
I also share that just because you have a nodule doesn’t necessarily mean I should biopsy it. There are other options. I go over the things that I look for in a suspicious nodule and what makes me more inclined to do a biopsy.
AMA: What is the biggest challenge in determining when a patient needs a biopsy?
Dr. Salat: We’re pretty good at determining which nodules are benign and which are concerning. But there is also a whole ocean of indeterminate nodules. Those are the ones that keep us up at night. There’s not enough evidence that you can just ignore them, but there’s also not enough evidence that you should act on them. It's a gray zone. But this has been an area of innovation.
AMA: What are some of the most exciting innovations when it comes to assessing lung nodules?
Dr. Salat: Ten years ago, the only option for patients who had a high risk of lung cancer, and a suspicious lung nodule was either surgery or a high-risk needle biopsy. Recent comparison trials have found that robotic bronchoscopy is much safer, with about a 3% chance of collapsing the lung compared to more than 20% for a needle biopsy.
These studies were published in the last two years. They confirmed what many of us had already experienced in practice, and now we finally have the data to support it. Innovation runs much faster than data.
AMA: What other technological advancements are having an impact on lung cancer detection and treatment?
Dr. Salat: There is excitement around how to use technology like blood-based testing, saliva testing, AI and machine learning to capture high-risk intermediate lung nodules early, when they are smaller and easiest to treat.
AMA: How are you using AI in lung cancer detection and treatment?
Dr. Salat: There is exciting research coming out in breast cancer where you can see a small lesion in the breast, and AI can determine the probability of it being malignant.
We’re beginning to see the same thing with lung cancer to determine if a lung nodule is on the spectrum of benign to malignant based on the radiological characteristics.
All the computations, analysis and risk stratification are driven by AI. There are large data sets and algorithms that are based on how a nodule looks on the CT scan. And there are large language models, or LLMs, based on thousands of scans that are built to detect whether the nodule could be consistent with malignancy now or in the future. These are exciting developments.
AMA: How do you combine the AI outputs with your clinical training and expertise to make patient care decisions?
Dr. Salat: For biopsies, I truly believe that just because you can doesn’t mean you should. Otherwise, you would be doing biopsies on patients left and right. At the same time, you have to balance that with watchful waiting as well.
AI isn’t going to stay up at night wondering whether we should do a biopsy on a patient. AI can just say it’s X, Y or Z. So, as a doctor, you have to make sure AI does not replace your clinical judgment. It can just help enhance it. That’s why this job isn't easy, and that's why AI won't replace us anytime soon.
AMA: What do you wish you had known before going into this specialty?
Dr. Salat: The training teaches you to get it right before you move—to be certain, to wait for the evidence. That discipline saves lives. It also slows innovation.
Medicine needs both instincts. Learning to hold them in productive tension is one of the hardest and most important things about this work.
AMA: What advice do you have for medical students considering pulmonology?
Dr. Salat: If you want a specialty where you are using your hands, your mind and your judgment every single day—and where the technology is evolving fast enough that your training is just the starting point—pulmonology and advanced bronchoscopy is a remarkable field right now.
The procedures are becoming more sophisticated, the clinical questions are genuinely interesting, and patients need you. Come ready to keep learning.
