In cardiology, every minute counts when it comes to receiving, analyzing, and acutely diagnosing patients. Early diagnoses and triage are essential in a cardiology workflow and can save lives. Unfortunately, in an emergency, everything competes for a physician’s time and attention.
Artificial intelligence (AI) enables speed and supports cardiologists. By enhancing early detection, helping doctors narrow down possible disease states, and decreasing false activations, AI can change the clinical course of a patient’s treatment for the better.
With the Rise of AI, What IP Disputes in Healthcare Are Likely to Emerge?
Munck Wilson Mandala Partner Greg Howison shared his perspective on some of the legal ramifications around AI, IP, connected devices and the data they generate, in response to emailed questions.
However, many physicians are skeptical of AI. They may believe it will undermine their judgment, independence, and autonomy, and that it foreshadows their replacement. I believe the implementation of AI should not be feared; based on the advantages it confers to physicians, it should actually be welcomed.
AI is a tool that can be used in both hospital and clinical settings to promote accuracy, efficiency, and treatment effectiveness. With the integration of AI into an imaging workflow, care can be triaged automatically to ensure the right physicians are notified of and weigh-in on patient data at the right time. Rather than removing the physician from the equation, AI gives clinicians higher quality data and confers on them the ability to generate better insight into disease states.
The incorporation of AI into existing cardiology care workflows allows the flagging of critical cases such as heart attack, ST-elevation myocardial infarction (STEMI), and atrial fibrillation (AF). In the United States, between 10,000 and 50,000 heart attacks are missed per year at emergency departments (EDs). The most common cause of inappropriate activation was due to the misinterpretation of electrocardiograms (EKGs) for STEMI criteria often due to the subjective interpretation by ED physicians or emergency medical technicians.
Unfortunately, these false activations can have unintended consequences. They can result in physician and staff fatigue, as well as worries that teams don’t take the activations as seriously over time. When this happens, both clinical and financial costs may be incurred.
A Personalized Approach to Medication Nonadherence
At the Abarca Forward conference earlier this year, George Van Antwerp, managing director at Deloitte, discussed how social determinants of health and a personalized member experience can improve medication adherence and health outcomes.
When I treat heart attacks, there can be a long delay between the time the patient gets an EKG to when I can read an EKG accurately on a device. I am not alone. Despite the development of many different systems and software, most hospitals still struggle to get an EKG in a HIPAA compliant manner to a cardiologist to review quickly. National guidelines from the American College of Cardiology and American Heart Association state that hospitals treating STEMI patients with emergency percutaneous coronary intervention (PCI) should reliably achieve a door-to-balloon time of 90 minutes or less.
A long delay can push the treatment window past this guideline and may impact patient outcomes. Recently, AI-assisted software has been shown to detect STEMI with a high diagnostic accuracy rate, making it possible to minimize delays in contact-to-treatment times for patients with ST-segment elevation. Like a compass, AI can point the physician in the right direction of care, increasing efficiency and decreasing overall healthcare spend.
Similarly, AI gives physicians precise guidance on who needs intervention. For example, when a primary care physician hears a murmur, they might ignore it, or they might order an echocardiogram. As there is no standardization in a primary care doctor’s physical exam assessment, detection is very important at this stage and an EKG read by AI that suggests aortic disease or mitral disease can help predict the need for further testing in an appropriate and efficient manner. Alternatively, physicians ordering unnecessary tests waste both time and money for patients and the health care system.
As a cardiologist, if I order a cardiac magnetic resonance imaging (MRI), it can take weeks after the study is complete to get the report back. Comparatively, AI analysis of cardiac MRI data may also be used to expedite the evaluation to minutes once the images are completed. It is clear that AI has the power to expedite care, benefiting patients.
For complex clinical cases in which multiple confounding factors are present and multiple diagnoses are under consideration, this may prove to be extremely helpful in saving time. Similarly, AI-based ECG analyses can improve diagnosis and patient management. AI diagnostic performance is already close to that of experienced cardiologists.
AI also furthers collaboration and analysis, fully informing discussions about treatment when questions arise. Stroke is one of the most dangerous and life-threatening diseases, but the life of a patient can be saved if the stroke is detected during early stage and a multidisciplinary approach is taken, with various specialists quickly delivering effective care. A multidisciplinary stroke unit team consists of doctors, nurses, physiotherapists, occupational therapists, speech and language therapists, and psychologists and may include input from pharmacists, radiologists, neurologists and geriatric medicine physicians.
The simultaneous activation of the appropriate care providers based on AI-backed diagnosis results in faster treatment time, which translates to better outcomes. Additionally, high-resolution images provided by AI-based software can enable physicians to easily obtain input from colleagues and experts in their field, anywhere in the world.
Overall, AI is an invaluable tool in hospital and clinical settings promoting diagnostic accuracy, workflow efficiency, and treatment effectiveness. AI gives clinicians the ability to access higher quality data and deliver potentially better insight into disease states. As more data become available and more sophisticated algorithms are developed, AI will play an increasingly important role in the diagnosis and management of cardiac conditions and other diseases. The integration of AI into cardiology workflows can help triage care automatically, ensuring the right physicians weigh in on patient data at the right time which may potentially improve clinical outcomes for all patients in need.
Photo: Sylverarts, Getty Images
Jonathan A. Aliota, MD, FACC, is an interventional cardiologist based in Houston, TX, and Cardiology Fellow at the Texas Heart Institute, and a consultant with Viz.AI. He is affiliated with multiple hospitals, including Houston Methodist Sugar Land Hospital and Houston Methodist Continuing Care Hospital. He earned his medical degree at Tulane University School of Medicine and has been in practice for more than 20 years.
