Is AI-assisted lung cancer diagnosis suitable for your hospital?

Is AI-assisted lung cancer diagnosis suitable for your hospital?

Lung cancer is the leading cause of cancer deaths worldwide, with approximately 1.8 million people dying from the disease each year. Most patients are diagnosed after the appearance of symptoms and progression of the disease to an advanced stage (stage III or IV), which explains the current worldwide five-year survival rate of only 20%. In contrast, the survival rate for small lung tumors treated at stage 1A reaches 90%. This significant difference highlights a critical need for diagnosing and treating lung cancer at the earliest possible stage.

One of the best opportunities to diagnose presymptomatic small lung cancers earlier is presented by the two million patients in the United States each year who have a incidentally identified pulmonary nodule during chest CT scans ordered for other reasons, such as during an emergency visit or after a cardiac event.

Current treatment guidelines call for one- to two-year follow-up to determine whether a nodule is cancerous. However, more than 60% of these patients do not receive guideline-recommended follow-up, severely limiting opportunities for early intervention and treatment. Patients who receive the recommended follow-up often require multiple imaging tests and biopsies, and sometimes unnecessary invasive procedures, such as surgical biopsies and lung resections, before reaching a definitive diagnosis.

Several factors contribute to this situation:

1. Interrupted fulfillment workflows. As noted earlier, a patient may receive a CT scan of the chest for a variety of reasons unrelated to a lung problem. During review of the scan, the radiologist notes that a pulmonary nodule is present and recommends follow-up by the patient’s primary care physician (PCP). However, at this time, this is a minor and non-urgent issue for this patient, so the care team may not alert the care team appropriately to manage the nodule. Also possible: the PCP evaluates the radiological report as non-critical and does not inform the patient. It is important to note that there are standard of care and legal liability issues associated with both scenarios.

2. Incidental screening diagnoses may not get the attention they deserve. Regionally, clinicians may be aware that a significant percentage of the local population has completely benign pulmonary nodules. And it’s true: 95% of those modules will remain benign. Therefore, when the patient’s PCP is informed of an incidental diagnosis, he may be hesitant to prescribe a treatment that involves a six-month course of CT scans – which are expensive and may unnecessarily alarm the patient.

3. The high cost of seeking a definitive diagnosis. It is widely accepted that nearly one-third of all CT scans that include part of the lungs describe an incidentally detected pulmonary nodule. Managing these patients with nodules can present enormous resource challenges in scheduling appropriate follow-up care. The larger the health network, the greater the challenge.

4. Low ROI. Implementing a workflow without automation to properly manage incidental lung nodule alerts is expensive and has a low ROI. Most hospitals are therefore reluctant to implement a program to diligently review lab notes from all tests. Clinical teams are already very busy, so allocating resources to screen benign nodules with conventional manual processes that require additional full-time staff is a low priority.

Scale up: the cost-benefit equation is changing

Recent advances in artificial intelligence (AI) are changing how these decisions are calculated. For example, an AI-based platform applies natural language processing (NLP) automation to instantly read and evaluate any radiology report, then identify and track patients who must receive special care. In addition, the system assigns a lung cancer prediction score to the nodes of interest, which stratifies patients and aids in accurate diagnosis. This, in turn, supports better clinical decision-making.

The potent combination of NLP- and AI-assisted diagnostic tools represents a viable solution for many healthcare systems, enabling the treatment of more early-stage lung cancers without increasing the workload of clinical teams. And by getting to the right diagnosis sooner, hospitals can also minimize unnecessary invasive biopsies.

Is AI-assisted lung cancer diagnosis suitable for your hospital?

Given the importance of early diagnosis, hospitals must implement a plan to screen for and manage incidental pulmonary nodules – to avoid reputational risk and save more patient lives. When evaluating your course of action, your clinical teams should ask the following questions:

1. In the past year, how many nodules were accidentally identified in your health care system?

2. Have they all been screened and what procedures are in place to recommend a path of care?

3. How many patients were lost to follow-up?

4. In 2023, if we screened and treated significantly more nodules adequately, could we do so without adding resources and staff?

If you can’t easily find any of the information above, it’s time to re-evaluate your approach. It is very likely that you have a serious problem that needs to be resolved.

About Ryan Hennen

Ryan Hennen is vice president of US sales for Optellum. He has more than 20 years of consulting experience with major healthcare IDNs while helping to deliver enterprise healthcare solutions. Ryan has experience in imaging, oncology, value-based care, population health, clinical decision support, AI, NLP and machine learning. Contact Ryan at [email protected] and LinkedIn.

References

1. Optellum projections based on Gould MK, Tang T, Liu IL, Lee J, Zheng C, Danforth KN, Kosco AE, Di Fiore JL, Suh DE. “Recent trends in identifying incidental pulmonary nodules,” American Journal of Respiratory and Critical Care Medicine, 2015 Nov 15;192(10):1208-14

2. Pyenson BS, Bazell CM, Bellanich MJ, Caplen MA, Zulueta JJ. “No apparent investigation for most new indeterminate pulmonary nodules in US commercially insured patients”, Journal of Health Economics and Outcomes Research, 2019;6(3):118-129.

Is AI-assisted lung cancer diagnosis suitable for your hospital?

Leave a Reply

Your email address will not be published.

Scroll to top