Health Insights

“So lung cancer is really important because it is the leading cause of cancer-related death in adults,” Dr. Kapur shared. “Particularly in our region,” she said of the Greater Cincinnati area. “We actually have more lung cancer than the country overall.”

As a cardiothoracic radiologist, she knows that timing can mean the difference between a patient’s life and a much more uncertain future. In her practice, she’s witnessed the emotional highs of early-stage diagnoses and the devastation that comes with late detections. But recently, she’s become an advocate for something she believes could revolutionize outcomes: using AI-assisted CT scan technology.

“Once you are in a science field, you know that you're learning for life,” Dr. Kapur shared. “How can I do more?” she reflected as she considered the challenges  of lung cancer detection.

The AI tools she’s implemented, such as the ClearRead system, have transformed how she approaches lung cancer screening. These advanced programs don’t just make the process more efficient; they improve accuracy and help radiologists spot what could otherwise be missed. “We are the first institution in the region using this technology to read more exams and to be able to find more of these nodules and help our patients.”

This article explores why early lung cancer detection matters and how Dr. Kapur’s innovative use of AI in medical imaging is changing the local landscape for lung cancer patients.

Why Early Detection Matters

“Lung cancer screening addresses this problem of finding lung cancer early,” Dr. Kapur said stressing the importance of CT Lung Cancer screening as a preventative tool rather than until you feel unwell to get a CT. She shared that only four percent of the eligible population nationally is undergoing screening for lung cancer while screening rates are higher in our region at seven percent.

“If a patient comes to us after they’ve developed lung cancer or they have symptoms,” Dr. Kapur gave as an example. “It’s a late-stage cancer. For most of these patients, the survival is poor. In the range of only 20 to 30 percent of people alive five years after they’re diagnosed.”

“So if we identify these patients who are at risk and we put them in annual screening, meaning they come in once a year to get a low-dose chest CT scan, most of those cancers, almost up to 90 percent we identify, are early stage and are curable.”

The Role of AI in CT Scan Analysis

For Dr. Kapur and her peers, the integration of AI has shifted how they approach the demanding task of reading CT scans. In the past, the sheer volume of slices in a single scan—often hundreds per patient—posed an enormous challenge. “What that means is that these things are really hard to find. They are a needle in a haystack. And if we have to do a good job. I need to either spend more time on each scan or I need to have more people reading those scans. But I have neither of those. So what do I do? I take the help of technology, which is out there and it helps me avoid the repetitive tasks.”

The ClearRead technology narrows down potential problematic nodules by clearing up all normal structures except the nodule from the image, thus removing "visual clutter" and allowing a radiologist like Dr. Kapur to go from focusing on dozens of spots to a handful of spots that truly need attention. The AI software can review scans in a matter of minutes. Despite this, Dr. Kapur calls the AI tool her “second reader” with staff always having the final look.

“This technology helps us find more nodules, helps us read faster, and is able to find all kinds of nodules,” Dr. Kapur said, noting that the AI technology thuse cuts down on time while increasing precision.

ClearRead’s unique vessel suppression capability allows radiologists to focus on the areas that matter most, enhancing the conspicuity of lung nodules.

“The first day the technology came into our department, I was sold because I had seen it in action, but my colleagues were just floored,” Dr. Kapur recalled when the AI software launched as the new standard for lung cancer screening at the University of Cincinnati Cancer Center in February 2024. “They were like, ‘This thing is unbelievable.’ How it finds all the nodules every time it measures them, it shows you what they are and what consistency they are, and then puts them in the report. I couldn't have asked for something better.”

This kind of assistance is invaluable. AI augments the radiologist's skill, ensuring that fewer potentially malignant lung nodules go undetected. For Dr. Kapur, this means she can approach each reading with greater confidence, knowing that her sensitivity to finding cancer has been bolstered. She points to data that back up her experience: ClearRead has been shown to improve the accuracy of radiologists, raising sensitivity from 64.5% to 80%. “It's been a game changer,” Dr. Kapur expressed with both excitement and gratitude.

ClearRead Technology Explained

The inner workings of ClearRead are impressive, rooted in cutting-edge machine learning and deep training with thousands of varied nodule cases. This has made it adaptable, even across different CT scan machines and settings. This adaptability is crucial for cancer centers like the University of Cincinnati Cancer Center that handle a broad range of cases and technologies.

What sets ClearRead apart is its precision in detecting nodules as small as 5 mm. In the past, these small findings could be easily overloofked, buried under a labyrinth of vessels and other normal structures. ClearRead minmizes these obstacles, and every nodule, whether solid, part-solid, or ground-glass, is assessed with improved clarity. This technology even provides automatic measurements for detected nodules, including their type, volume, and dimensions, supporting more accurate assessments.

Impact on Workflow and Efficiency

In high-stakes fields like radiology, time is precious. Dr. Kapur emphasizes that the AI ClearRead software doesn’t just help identify nodules; it improves workflow without adding more steps. She says this helps her and her team read more scans faster, reducing their time staring at a computer screen and focusing on one scan. Instead, they can work faster and reach more patients.

“We're able to read more scans. We are able to focus on the things that matter. We are able to focus on that needle in the haystack, that spot that matters. And we are able to find more cancers, early and thus help our patients,” Dr. Kapur said.

Dr. Kapur shared how, “the [University of Cincinnati College of Medicine] residents are transformed because they see this technology and they're like, ‘This is the part that is most tedious.’ And now the tedious part is out of the system.”

This capability has improved reading times by 26%, enabling more patients to be seen and diagnosed promptly. Additionally, this software has been shown to help find 30% additional nodules that may have been missed, and some of those may have been cancerous. This behind-the-scenes efficiency translates to better patient care, allowing physicians to spend more time discussing findings and next steps with their patients instead of being buried in image analysis.

Patient Outcomes and Future Implications

Dr. Kapur’s belief in AI is driven by one core truth: it saves lives. Early detection through enhanced lung cancer screening means patients are more likely to receive a diagnosis at a treatable stage.

“I definitely feel blessed to be able to kind of see this transformation in front of my eyes where we took so many things for granted,” Dr. Kapur reflected. ‘Well, okay. Well, I guess it's fate. I was going to get cancer.’ But now we can say no. You don't have to. You know, there is there is hope for us,” she emphasized.

ClearRead's improved lung nodule detection rates directly contribute to better prognoses for patients who might otherwise be diagnosed too late.

Dr. Kapur hopes that lung cancer screening will continue to expand, incorporating AI and additional cancer risk factors. Through the new screening tools and future innovation, she hopes patients can continue to find better outcomes and cures for their own journeys.

“We can catch these things very early and maybe in the future, you know, think about, making a change before this thing comes to fruition. So I think we are living in an era of explosive change in the way we address disease,” Dr. Kapur said.