AI-Powered Stroke Centers: Redefining Patient Outcomes

Why New Technologies for Stroke Care?

During a stroke, we lose anywhere from 30-32,000 brain cells every second. A 10-minute delay in treatment could cause the loss of 19 million brain cells. According to the Stroke Awareness Foundation, in the United States alone, strokes are a leading cause of death and disability, affecting 1 out of 20 people and killing around 140,000 people annually. Every four minutes, a person dies due to this illness, with someone experiencing a stroke every 40 seconds.

The role of technology in stroke prevention, detection, care, recovery and rehabilitation is constantly growing. Gadgets, devices and apps are being used everyday by both healthcare providers and patients in order to monitor cardiac health. In particular, the evolution of Artificial Intelligence (AI) has facilitated huge advances in stroke care and brain aneurysm care.

Why AI for Stroke Care?

Prior to AI, a CT scan would be performed after which the treating physician would seek out the on-call radiologist. Typically, this process would involve many phone calls, triaging and trying to figure out efficient follow-up. By the time a diagnosis was reached, hours (rather than minutes) had been lost. 

AI is creating an enormous difference to help make up for lost time, assisting radiologists while supporting care coordination efforts. AI is already facilitating change in many areas of our lives, and its evolution is now reaching healthcare, helping overburdened physicians to maintain control and be in charge. That’s because AI ensures speed and efficiency in patient care – essential, life-saving capabilities in the treatment of strokes – and delivers a series of important benefits to healthcare:

  • Automation: AI is always on, running in the background and analyzing each exam.
  • Prioritization: AI prioritizes findings of positive cases in the radiologist’s worklist, reducing stroke detection time and time to treatment by notifying the on-call stroke interventionalist, thus enabling cross department activation and collaboration. 
  • Enhanced Workflow: AI streamlines hospital workflows by notifying relevant physicians while providing communication tools and image viewing within one mobile platform. 
  • Communication: AI allows all pertinent information and images to be located in one place, and for streamlined, secure communication between all relevant personnel. At the push of a button, it’s possible to get hold of the ER physician, the critical care physician, activate the stroke team, and then go into the operating room seamlessly. This helps patients because it speeds up the entire treatment process, and also helps physicians as it reduces friction points while time is of the essence. 
  • Transfer Management: AI-enabled synchronized care gives transfer managers the means to quickly and efficiently effectuate the transfer of patients between hospitals, including sharing images and EMRs. This is of particular benefit within the hub & spoke model, where stroke patients are transferred to the “hub” facility containing the interventionist team.
  • Confidence in Technology: AI must be accurate with low false-positive rates in order to give physicians the confidence they need to adopt the technology.

How Should the Adoption of AI be Approached?

The evaluation and implementation of AI is a new process – from raising awareness amongst stakeholders and reviewing options to selecting the most suitable vendor and starting your journey. Here are three important tips:

  1. Work with a Team: Evaluating and implementing AI is a new process which cannot be done successfully without having an expert tech person (IT, informatics) working with a clinical expert. You’ll also need AI champions in the clinical realm and a Transformation & Digital officer.
  2. Safety & Quality: Will the proposed AI solution improve safety and quality? Does it have a proven, high-level of accuracy? Will it improve, streamline and speed up the treatment of stroke patients? Will it reduce mortality rates from strokes? Does it automate routine, mundane tasks and thereby lower costs? These are examples of important questions you should be constantly asking.
  3. Partnership: Deploying AI is a long-term relationship. To be successful, it requires a partnership with a vendor that you can trust; a vendor with published case studies, FDA clearances and/or CE markings and a proven track record of improving workflows.

Benefits Observed with AI in Stroke Care

At a recent webinar entitled The Future is Here: AI-Powered Stroke Centers, Dr. Eric Eskioglu of Novant Health and Dr. Brian Mason of Carle Foundation Hospital discussed their experiences with diverse AI tools, noting the following important benefits that they observed in their healthcare facilities after adopting AI:

  • Increased safety and quality.
  • Less penalties paid to Medicare because of readmissions (because there are notably less readmissions). 
  • Less siloed data thanks to  better communication and accessibility of data between radiologists, physicians and specialists.
  • Equal and equitable access to healthcare, because AI extends the reach of the technology to rural areas, enabling the provision of fast and immediate, top-quality healthcare – something that was not previously possible. 

“AI not only provides equity, but also lowers the cost of care, because we’re now treating them [patients] faster, getting them back home faster. And by doing that we’re lowering the access entry for patients that normally could not afford healthcare.”
Dr. Eric Eskioglu, EVP, Chief Medical Officer & Scientific Officer, Novant Health 

  • High accuracy rate of positive findings.
  • Easy, streamlined communication that eliminates many previously time-consuming challenges and frees surgeons to develop better techniques for neuro stroke care.
  • Support in making quick, informed decisions.
  • Getting to patients faster.

Where is AI Going to Progress in the Future?

Patient data is doubling every 72 days, and with that comes the increasing challenge for physicians and radiologists to play catch-up. Add to that the severe shortage of physicians and healthcare personnel because of burnout and retirement (60% of U.S. physicians in internal medicine are above the age of 55). And don’t forget that the number of patients is increasing. AI has to be implemented to help resolve these issues. It’s going to be an essential data collection tool. It’s going to be a way of life in healthcare. 

Hospitals and healthcare providers will choose a full AI platform, rather than AI-based point solutions. And the platform will not just be used for radiology, but also to coordinate complex medical care and advanced therapies. AI – with the capabilities of natural language processing, data collation, trend identification and prognosis – will allow physicians to spend more time with their patients and get a better diagnosis for treatment.

“I think that we’re going to get to a point where we’re going to need AI to help collate data and present it in a better fashion to the physician.”
Dr. Brian Mason, Senior Neuroendovascular Surgeon / Associate Professor, Carle Foundation Hospital

“AI is never going to replace us as physicians. But physicians who use AI are going to be so much more effective and so much more efficient both in diagnosis and time spent with the patient, that they’re going to replace physicians who resist AI. I can give you 100% certainty that is going to happen.”
Dr. Eric Eskioglu, EVP, Chief Medical Officer & Scientific Officer, Novant Health 

Learn more about AI-Powered Stroke Centers.

Explore the Latest AI Insights, Trends and Research