Future of Thoracic Care Is Identifying Lung Cancer Early: Samir Shah, MD, MMM, FACR

Samir Shah, MD, MMM, FACR, spoke about his collaboration with University Hospitals (UH) Cleveland to incorporate more artificial intelligence (AI) tools that could help identify early nodules of lung cancer. The future of thoracic care could rely on these tools to help improve prognosis in lung cancer.

This transcript has been lightly edited for clarity; captions are auto-generated.

Transcript

Can you share how qXR-LN works and its significance for thoracic specialists?

Yeah, this is a very interesting collaboration, particularly with Dr [Amit] Gupta, who's the head of cardiothoracic radiology [at University Hospitals]. One of the things [is] that his interests are 2-fold: 1 is lung cancer, and 2 is AI. He's an informaticist and an expert in AI, and he was a great collaborator.

One of the things we did, or are doing right now at UH Cleveland, is we're actually looking at how many more lung cancers we're finding. And the way we're doing that is we're actually blinding the radiologist to the use of AI, and we're actually turning it on 1 day, turning it off the next day. It's actually a beautifully designed program. It's in use clinically throughout the UH Cleveland network, and the results basically are told to the radiologist afterwards, and then the radiologist will make a correction or addendum to their study and communicate that finding. That's how it's being used today.

The real problem is that radiologists don't have time like they used to have 20 years ago. We're overloaded with studies. Our focus is kind of on CT, MRI studies, ultrasound studies. X-ray's kind of fallen a little bit by the wayside, so we read them very quickly and not with the same detail. And it takes detail to find a little nodule. Some of the nodules we find with AI are actually better than a human radiologist can find because we can find nodules of a certain size, but the AI can find smaller nodules with more sensitivity and specificity than even a human.

How do you see the future of thoracic care evolving within the next couple of years?

I think that...the most important thing we can do is stage-shift lung cancer. Just for general awareness, right now lung cancer is the number 1 killer in the US as a cancer and in the world. Lung cancer will kill more patients than breast cancer, prostate cancer, [and] colon cancer combined. And this is really, really horrible.

The future of thoracic care, it's going to involve multiple advanced technologies. It's not just going to be computer vision and generative AI, we're going to need things like liquid biopsy and different types of screening, because it's not just smokers anymore that are getting lung cancers. For example, Asian females are one of the highest-growing numbers of lung cancer, and there's no way to find those patients unless we apply technology to the entire population.

That's really the future that I see, is that we're going to catch cancers earlier in patients who don't typically fall under the screening guidelines. The typical patient has many years of smoking history and comorbidities like emphysema. We need to find the patients that are not under those specific guidelines and just might be walking around with a small lung cancer [tumor] that doesn't cause any symptoms, and we can find it early and treat it. The way they're treating these things now, the surgeries are minimally invasive. There are new technologies like burning the small tumor with microwave[s] or some sort of technological freezing of the tumor to basically kill the tumor in a noninvasive way. This is all advancing. I'd probably say that thoracic is becoming the most advanced specialty in all of medicine.