As we observe National Cancer Prevention Month this February, medical professionals consistently highlight several preemptive measures, including lifestyle changes and regular screenings. Research shows that up to 50% of cancer cases may be preventable, but for millions of Americans, they face significant hurdles to vital measures like early detection screenings simply because of geography. Innovators are helping to bridge this gap.
According to the Centers for Disease Control and Prevention (CDC), “Rural residents are more likely than urban residents to suffer and die from some of the deadliest but most preventable cancers.” Roughly 20% of Americans live in rural areas, yet these communities are significantly underserved, lacking access to quality resources for cancer prevention, screening and treatment, which are often concentrated in urban centers. When the nearest specialist or diagnostic center is a three-hour drive away, routine, preventative care can easily become too difficult to maintain.
As we have highlighted in the past, artificial intelligence (AI) is being integrated across the healthcare industry, including in medical imaging and breast cancer diagnostics, and it is well-positioned to help close this rural access gap. AI is making it possible to democratize world-class medical expertise, delivering it directly to rural clinics and patients.
The keen prediction capabilities of AI tools make them an extremely advantageous addition to supplement preexisting cancer prevention methods. By analyzing electronic health records (EHRs), rural health care providers can use AI to identify which patients are at a higher risk of developing certain cancers. In a 2022 study conducted in the UK, five different machine learning techniques were compared against a widely used primary care cancer risk-assessment tool for oesophago-gastric cancer. The machine learning algorithms outperformed the current risk-assessment tool with a higher overall accuracy and the ability to identify 11-25% more cancer patients.
At Carnegie Mellon University, researchers recently developed one of these non-invasive tools to prescreen for cancer using only patients’ EHRs, called CATch Cancer Early With Healthcare Foundation Models (CATCH-FM). Functioning similarly to chatbots, preliminary tests showed that the technology correctly flagged 50% (among patients with no prior cancer diagnosis) to 70% (among patients with prior cancer diagnoses) of patients as high risk who were, in fact, later diagnosed with lung, liver and/or pancreatic cancer. Integrating AI into predictive risk modeling allows local healthcare providers to help high-risk patients sooner and better, shifting later rural detection rates from reactive treatment to more opportunities for proactive prevention, and hopefully, one day, lower death rates.
The evolving technology is also empowering rural primary care to better assist with screenings and early detection, as a local physician or nurse practitioner can now use AI-assisted diagnostic tools, trained on millions of medical images and data points, to evaluate patients. For example, medical professionals can use DermaSensor, an FDA-cleared handheld device designed to assist non-dermatology expert physicians. The device employs Elastic Scattering Spectroscopy (ESS), a visual tissue sampling technique that does not require surgical biopsy, enabled with an AI algorithm to assess skin lesions for cancer.
In a study published in the Journal of Primary Care & Community Health, there were several key findings when comparing visual assessment only rates to those assisted by the device’s output, including: diagnostic sensitivity increased significantly by around 10.6%, referral sensitivity increased significantly by 9.4%, and physicians reported a higher confidence in their management assessment by 16.6%.
Technological advancements are helping ensure that access to quality, preventative healthcare will not be dependent upon one’s zip code. To work towards this beneficial and potentially life-saving democratization of care, we must have a policy environment that fosters innovation rather than stifles it. This National Cancer Prevention Month, we should consider the benefits that AI integration into healthcare offers, particularly in making preventive care a reality for all Americans, no matter where they call home.
Image via Unsplash.