AI Insights: Community Hospital

Community Hospitals make up 85% of all hospitals in the United States. The American Hospital Association (AHA) defines community hospitals as “all nonfederal, short-term general, and other specialty hospitals.” AI will have an immense impact on the delivery of care and management of community hospitals. We examined the topic in detail at the World Medical Innovation Forum. We have also added an exchange with one of the nation’s top telestroke experts Dr. Lee Schwamm.


WMIF Panel Video

Community hospitals are the largest sources of patient care in the U.S. As such, they represent a frontier in the transformation of health care. How are these organizations using AI and digital technologies to drive transformation? What are the distinctions from academic medical centers? This session will address these and other topics that impact community hospitals.


First Look: Deep Learning for Glaucoma Detection: Nazlee Zebardast, MD, Instructor, Ophthalmology, MEE, HMS

Glaucoma is the leading cause of irreversible blindness worldwide, with an estimated global prevalence of 3.5% in persons aged 40 to 80 years old and affecting approximately 64 million people worldwide as of 2013. Glaucoma is an ideal disease to screen for as early treatment of glaucoma has been shown to delay glaucoma progression. Watch Nazlee Zebardast, MD discuss glaucoma screening in a potential community setting.


D12: Number 9- Revolution in Acute Stroke Care

Stroke is a major cause of death and disability across the world and a significant source of health care spending. Each year in the U.S., nearly 800,000 people suffer from a stroke, with a cost of roughly $34 billion. AI tools to help automate the diagnostic journey of ischemic stroke can help determine whether there is bleeding within the brain — a crucial early insight that helps doctors select the proper treatment. These algorithms can automatically review a patient’s head CT scan to identify a cerebral hemorrhage as well as help localize its source and determine the volume of brain tissue affected.


Lee Schwamm

Q&A with Lee Schwamm, MD, Director, Center for TeleHealth and Exec Vice Chair, Neurology, MGH; Professor, Neurology, HMS

  1. Community hospitals have been dubbed the “heart and soul” of healthcare, as they’re often the single source of care for millions of people across the country. What role do you see AI and telehealth playing in delivering better care to these patients?
    • Dr. Schwamm (DS): Community hospitals have the opportunity to perfect the delivery of high-quality care at lower cost. This is incredibly important to the delivery of healthcare. We are trying to create systems to reduce variation in cost and quality as AMCs transfer care protocols to their consolidated networks of community hospitals. This is where telehealth and AI have an important role to play. Both can capture expertise and knowledge to package and deliver information effectively. With telemedicine, you are delivering synchronized care in real-time, so you don’t have to move patients or doctors around. Artificial intelligence is a way for knowledge to be stored in “virtual machines” and this intellectual content can be then applied at another location and at another time without that expert having to be available. This amplifies knowledge exponentially.
  2. What’s an example of this process?
    • (DS): An example of telehealth being used in the community hospital setting is telestroke, which is the use of telemedicine to assist ED providers in the management of patients who arrive with an acute stroke within the first few hours of symptom onset. For ED physicians in community hospitals, recognizing when acute stroke is occurring requires repeated exposure to patients showing these symptoms, yet stroke is a low frequency, high impact event representing only 5% of all ED visits and hospital admissions. This means that many healthcare providers are not exposed to these symptoms often enough to become adept at diagnosis. This is the perfect model for telemedicine since it connects stroke experts who are concentrated in specific locations with patients and ED providers who are widely distributed.
      Looking ahead, a way that artificial intelligence could add value is when EMS arrives on scene and tries to determine whether a patient is having a stroke. The EMT could enter into a smartphone the results of their findings and AI can use a series of decision trees and probabilistic models to help determine if the patient is having a stroke. This could also trigger a list of hospitals in area, and calculate the destination that would treat the patient in the fastest amount of time and ensure the best care based on individual hospital performance data.
  3. For physicians and caregivers in community hospitals, how can technology (and AI) help them grow as medical professionals and support them in delivering top-quality care and treatment?
    • (DS): is is an important issue as regionalized care is a double-edge sword.  Making expertise available remotely is useful for the patient, but if the providers are disengaged and see it as a way to defer care to another provider, they lose capabilities. Technology should be considered part of ongoing professional development and maintenance of certification. We need to be building capability and capacity in community hospitals, not diminishing them. Knowledge from neurology experts and other specialties can be shared, captured and coded into practice at community hospitals, increasing care capabilities everywhere.
  4. In general, do you see community hospitals embracing technology and AI, or is there an arc to adoption? What can help continue this momentum or accelerate it?
    • (DS): Once people can associate remote expertise as a complement to onsite expertise, acceptance increases.  As MGH TeleNeurology consultants were introduced for hospitals, it was difficult at first until people started using it. Now that people have given it a try, it’s becoming accepted and widely used.
  5. Do you foresee that telehealth will be used more in other care settings, including AMCs — will visiting the doctor become a thing of the past, and if so, when?
    • (DS):I don’t think seeing a doctor in person will ever be a thing of the past. However, there are better ways to do certain transactions. As the cost of healthcare shifts more to the consumers, consumers will make decisions based on their out of pocket costs. So, we need to consider how what we do creates value for everyone and not just value for providers or hospitals. Roughly 20% of in-person visits could likely be replaced by video. This will give more time for doctors to see new patients who need an in-person initial assessment. AI also gives doctors the opportunity to manage patients more effectively throughout their journeys. Specialists can be automatically reengaged if something is going wrong.
      Telehealth is both disruptive and innovative. We may see the best outcomes in areas we didn’t even imagine. We can and should “expect the unexpected”.

Quick Take Video from WMIF

Fabien Becker, PhD, CEO, Arterys discusses potential opportunities and challenges of harnessing artificial intelligence tools in a community hospital setting.