Could AI help heal rural healthcare?

In the vast rural areas of the U.S., where about one in every five Americans lives, the landscape varies widely—from Alabama's farmlands to the winding valleys of West Virginia, all the way out to Nevada's high desert. But all these diverse areas share a common healthcare crisis. People in rural America are more likely to die from heart disease, cancer, and other major illnesses than those in cities. And it's getting tougher to find care, with 191 rural hospitals having closed since 2005.

The problems are many and varied: not enough doctors, scarce special medical services, hospitals struggling with money, and other big-picture challenges. It’s not a one-size-fits-all problem. "The issues in Pikeville, Kentucky, for example, differ markedly just from those over in Mingo County, West Virginia,” says Dr. Phillip Polakoff, a Consulting Professor at Stanford and founder of A Healthier We, a non-profit dedicated to addressing issues in rural health. 

While there's no easy fix for the complex set of challenges in rural healthcare, AI is emerging as a potential change agent. Consider the AI advancements in medicine that have broken news in the last year: AI's newfound ability to grade neuroblastoma marks a major step forward in pediatric cancer care. In a recent study, an AI model, accessible via a phone app, outperformed clinicians in diagnosing pediatric ear infections. And AI's skill in analyzing eye scans is setting the stage for early Parkinson’s disease detection—years, potentially, before symptoms show up. Most of these AI advances focus on remote care and diagnosis. For underserved areas with limited access to specialists, innovations like these could be a lifeline. Beyond that, there's the possibility of AI potentially reducing "work about work" for doctors, creating patient narratives from a complex web of records.  

Polakoff is optimistic on AI—but cautiously so. “There are definite upsides to AI: from cleaning up electronic records to more personalized care to earlier diagnoses. But we’re going to need approaches that meld technological advancements with the most important thing: human touch.”

Here are five ways AI might make healthier outcomes possible across the rural US: 

1. Helping electronic health records deliver on their promise.  
Despite the switch from paper charts to electronic health records that accelerated beginning in 2009, U.S. healthcare providers often struggle with disorganized systems. (Who hasn’t been in an exam room, retelling their medical history to a doctor, while wondering why it isn’t already on the screen in front of them?) Dr. Ilana Yurkiewicz, a Stanford oncologist, delves into these challenges in her 2023 book, 'Fragmented: A Doctor's Quest to Piece Together American Health Care,' where she describes how doctors navigate through scattered electronic data to piece together a patient’s history.

AI offers a solution by making it easier to process vast amounts of data and identify relevant information, aiming to create a more coherent narrative from fragmented records. Yurkiewicz — another cautious skeptic—says that in a best case scenario, AI’s capacity for intelligent data analysis and organization could “clean up the messes of old tech,” enhancing how doctors access and interpret information for better decision-making.

That’s just the purpose of initiatives like MedKnowts at MIT and Beth Israel Deaconess Medical Center. By simplifying electronic health record interfaces and efficiently and automatically presenting relevant patient information, AI-enabled systems could minimize the time clinicians spend navigating through complex data.

Developed for widescale use, the advancements could be transformative. “Especially in settings where there are shortages, you want to make sure that the doctor is working at the top of their license,” says Yurkiewicz. “You don’t want a doctor’s time bogged down in clerical tasks that technology can and should do.”

2. More targeted and effective cancer screening. 
We’ve written about the potential of AI to analyze genetic data for better cancer treatment decisions. But AI is also showing promise in early disease detection and targeted testing. This approach could be crucial where resources are scarce.

Dr. Adam Yala, an Assistant Professor in Computational Precision Health at UC Berkeley and UCSF, explains the shift in strategy: “Cancer screening is more than just reading an image; it's about getting the right image at the right time. Right now, we tend to recommend the same screenings for everyone, like mammograms for all women past a certain age, regardless of individual risks. This can lead to late detections or unnecessary stress and procedures.”

Dr. Yala advocates for the use of AI to create tailored testing schedules, adapting to each individual’s risk factors. AI’s detailed analysis of medical images could significantly improve the accuracy of cancer risk predictions. In places with limited resources, like fewer MRI machines or specialists, AI's efficiency in screenings could mean more effective use of what's available. As Dr. Yala puts it, “The more limited the resources, the more crucial and impactful this kind of prioritization becomes.” 

3. Improving EMS outcomes.

Calling an ambulance in rural America can be a fraught experience. People in rural areas wait nearly twice as long for ambulances compared to those in urban areas; sometimes up to an hour or more. These delays are critical; even a single minute can increase the risk of fatalities.

But in the future, AI-driven systems could optimize EMS dispatch and routing, ensuring faster responses over rural distances. By analyzing emergency call patterns, AI could predict when and where ambulances are needed most, allowing for smarter deployment of resources.

AI's role in triaging emergency calls is already showing promise. Systems like Corti analyze speech to help dispatchers quickly identify urgent situations, such as heart attacks. AI is also improving triage by efficiently processing data faster than dispatchers can. Technologies like RapidSOS are linking data from smart devices directly to emergency teams. And new technologies are under development to assess car crash severity in real time, a crucial advancement for rural areas where timely response is even more critical.    

4. Broadening access to mental health care.

Mental health is a huge concern across the country, but especially in rural America: 7.7 million nonmetropolitan adults reported a mental illness in 2022, and 5.7% had serious thoughts of suicide. But the access-to-care gap is much wider outside of cities. In rural areas, there’s a shortage of mental health professionals, affordability issues, and a prevalent stigma associated with seeking mental health care in small communities.

AI offers a novel approach to bridging this gap. AI-powered chatbots could provide remote mental health support, crucial in areas where in-person services are scarce. These AI tools can offer immediate assistance, conduct preliminary assessments, and provide ongoing support. Though the space is moving quickly, concerns remain about biases in AI training models, potentially leading to misinterpretations in delicate mental health situations. 

AI is also at the forefront of a trend with remote patient monitoring—the use of devices that can track indicators linked to mental health conditions like sleep patterns, activity levels, and heart rates. Those insights could help a doctor understand the severity of symptoms for conditions like depression and bipolar disorder, even if they’re located hours away.  

5. Enabling sophisticated remote monitoring and more complex procedures. 

It’s likely that you took part in the telehealth boom during the pandemic. For many of us, it was the first time we took a video visit with a doctor. With the surge in AI’s capabilities, telehealth could encompass applications that go far beyond remote consultations. Take, for example, AI-enabled neural stimulators, which offer promising treatments for conditions like Parkinson’s Disease. The devices allow for remote management and adjustments, particularly helpful when a patient lives far away from their specialist. “It’s one of the developments in AI that I’m most excited about,” says Polakoff.

Surgery in rural areas could change, too. Technologies like the DaVinci surgical system—which increasingly relies on AI to predict complications and optimize techniques—could further enable complex surgical procedures to be performed in hospitals that lack specialized surgeons. Operations that were once confined to well-equipped urban centers would continue to become more broadly accessible to rural hospitals, bridging the gap. 

“AI, if it’s used judiciously and effectively, has a place in rural health sooner rather than later,” says Polakoff. “But it’s going to take funding, of course, and smart leadership—at both the top and the bottom.”