Skip to main content
Rural Health Information Hub
Rural Health Information Hub

AI-Powered Image Analysis

AI can help healthcare providers in rural communities, such as radiologists, by assisting with review of medical images, including X-rays, CT scans, MRIs, ultrasounds, and clinical photos. These images are commonly used in fields such as radiology, oncology, and dermatology.

AI-powered image analysis tools use deep learning models trained on large amounts of data — in this case, many different medical images that have been labeled by medical experts — to spot patterns. AI tools can automate parts of image analysis, meaning that healthcare providers can diagnose conditions more accurately and efficiently.

There are now over 1,000 AI and machine learning medical devices authorized by the U.S. Food and Drug Administration for use in healthcare settings, and many of these tools are designed to help with medical imaging.

The potential benefits of AI-powered medical imaging for rural healthcare include:

Faster image review and decision support, helping providers to quickly pull together information from many images, past cases, and clinical notes.

Support in settings with limited specialists, for example, in communities with fewer radiologists, cloud-based AI models can help review medical images and identify areas of concern.

Improved access through cloud technology, which does not require the purchase of onsite hardware, making AI image support more affordable and accessible to small, rural healthcare providers.

Easier consultation with offsite specialists, as cloud-based tools make it possible to share images and confirm findings with specialists who may be located hundreds of miles away.

Support for timely, high-quality care close to home, as cloud-based AI imaging tools can help fill gaps caused by workforce shortages of specialists and limited access to diagnostic equipment.

Examples of Rural Programs Using AI-Powered Image Analysis

  • Mercy, a large hospital system based in Missouri, uses cloud-based AI imaging to help doctors review scans such as CTs and X-rays across more than 50 acute care and specialty hospitals. The AI system does not replace the work of radiologists — instead, it helps them handle the large volume of images and identify patients who need immediate care.
  • Avera Health, which serves a large rural region across South Dakota and neighboring states, uses a shared, cloud-based imaging system to support radiology care. To improve quality and efficiency, Avera established a shared picture archive and communication system across its health system, overseen by a peer review committee. This approach has reduced medical costs, increased the number of images reviewed, and supported improvements to physician workflow.
  • Texas County Memorial Hospital, which serves rural southern Missouri, is using an FDA-approved AI tool to improve stroke care. The tool uses AI and machine learning to analyze CT scans, looking for signs of blood vessel blockages, and sends results to a remote stroke specialist in real-time. When a patient comes to the emergency room with stroke symptoms, staff conduct a CT scan that is automatically analyzed by the AI tool. The specialist is then prompted to read the CT results and recommend a course of action. This program has saved staff valuable time and improved patient care and outcomes.
  • Geisinger, a large nonprofit health system based in Pennsylvania, used 47,000 CT scans to develop a deep learning algorithm to spot signs of bleeding in the brain. The AI technology reviews each scan before a radiologist sees it and flags CTs that may show bleeding, helping doctors to diagnose patients faster, which can improve patient outcomes.
  • The Risk Underlying Rural Areas Longitudinal (RURAL) cohort study, and its RURAL ECHO ancillary study, is a federally-funded research project focused on heart and lung health in rural communities across Alabama, Kentucky, Louisiana, and Mississippi. The RURAL team uses mobile research vans to travel to rural communities and provides different services, including AI-based echocardiography. This AI tool uses deep-learning algorithms to provide step-by-step guidance to staff who are not imaging specialists so they can capture high-quality heart images.

Considerations for Implementation

Medical images are critical for diagnosing disease, and AI and cloud technology have the potential to help rural healthcare providers review medical images faster and more accurately. AI tools can help support providers by providing information and prompts so they can analyze large numbers of images more quickly and in real-time. However, this technology does not replace physician judgment. Today, most AI and machine learning devices are used to support humans; only a limited number of tools have been tested without human oversight.

When implementing these tools, rural healthcare organizations have noted the need for strong leadership and decision-making ability. Key challenges or concerns include patient privacy, data security, and provider trust and awareness. There are also concerns about AI tools trained on data that may not reflect populations served, highlighting the need to ensure AI tools used in rural practice are trained on patient data that reflects populations similar to those in rural or underserved communities.