Getty Images/iStockphoto

How AI has cemented its role in telemedicine

Many healthcare clinicians rely on AI when performing daily tasks and see benefits that outweigh the drawbacks.

For many years, healthcare providers have looked to technology for assistance and to minimize human errors. AI is now at the forefront of this effort.

Technology permeates so much of healthcare today that it's impossible to separate telemedicine from medical care in general. Many experts consider any medical interaction facilitated by or involving technology as telemedicine. While not every instance of telemedicine involves AI, the use of AI within the field of healthcare has expanded dramatically in recent years.

As the number of healthcare interactions that qualify as telemedicine have grown, so too have the number of patient touchpoints that involve AI, according to Elizabeth Krupinski, associate director of evaluation for the Arizona Telemedicine Program and director of the federally funded Southwest Telehealth Resource Center.

The breadth of use cases for AI in telemedicine has expanded and will continue to grow in the future. While experiencing this phenomenon, the healthcare industry has seen both benefits and drawbacks.

AI use cases in telemedicine

Many of the computer technologies and digital tools used by clinicians and patients have built-in AI capabilities, making AI commonplace in healthcare settings and health-related interactions. It's used in ICU command centers to analyze patient data and alert to crises. AI-driven tools monitor patients, both inside medical facilities and outside hospital walls. AI even helps clinicians triage patients, diagnose their conditions and plan optimal treatments.

Julius Bogdan, a vice president with the nonprofit association Healthcare Information and Management Systems Society, detailed broad categories of healthcare that incorporate AI:

  • Remote patient monitoring. AI -- specifically machine learning -- takes in and analyzes a patient's vital signs and alerts relevant parties to any abnormal readings. AI commonly analyzes data from blood pressure cuffs, heart monitors and other medical devices to watch for anomalies.
  • Patient diagnosis and medical image analysis. Using both individual patient data and larger sets of historical data, AI helps clinicians reach the most accurate diagnoses and accurately interpret the results of medical imaging.
  • Treatment plans. AI can personalize an optimal course of medical intervention based on analysis of a patient's unique profile.
  • Patient engagement. AI-driven technologies, such as chatbots, streamline services like providing information, scheduling appointments and handling intake prior to clinical visits.
  • Chronic disease management. AI can help monitor a patient, provide feedback to him or her and alert to early warning signs of disease progression.

Benefits and drawbacks of AI in telehealth

Medical providers reportedly have a favorable opinion of AI in healthcare. A 2019 report from MIT Technology Review Insights, in association with GE Healthcare, found that 75% of medical staff who use AI said it has enabled better predictions in the treatment of disease. Additionally, 78% of medical staffers experienced workflow improvements and 79% said AI has helped avert healthcare worker burnout. A majority of respondents also said AI lets them spend more time performing medical procedures instead of administrative and other such tasks.

AI in telemedicine -- and in healthcare generally -- helps do the following:

  • Divert time away from administrative tasks to medical care. General office tasks can take valuable time away from direct patient care. AI can perform these tasks instead.
  • Speed time to treatment. By collating, synthesizing and analyzing data from multiple sources in near real time, Ai delivers data-driven insights. Clinicians use this information to quickly and accurately determine the best course of action for their patients.
  • Extend the reach of medical care. AI-fueled technologies, such as remote patient monitoring tools, allow clinicians to treat patients in their homes as well as in rural locales where there are often few or no medical facilities.
  • Develop more personalized treatment plans. Algorithms analyze historical data as well as each patient's data to determine optimal medical interventions.
  • Manage chronic conditions and diseases. AI can create personalized treatment plans and continuously monitor patients following their regimens.

These benefits produce cumulative positive impacts in healthcare.

"AI can improve overall patient experience and patient outcomes through more accurate diagnoses, more effective treatment plans and speedier delivery of those services," said Amar Gupta, a research scientist at MIT's Computer Science and Artificial Intelligence Lab.

AI in telemedicine has some drawbacks and comes with challenges to its use.

There are integration issues. The MIT and GE Healthcare study found that 57% of respondents experienced challenges integrating AI applications into existing systems. Costs associated with AI adoption can also be a deterrent.

There are also data privacy concerns. Related to patient privacy are worries around accuracy, bias and reliability.

"One of the biggest challenges in a clinical setting … is that AI tends to be perceived as a black box and people don't trust black boxes. We're overcoming that, but it's still a barrier," Bogdan said.

The future of AI in telemedicine

Despite such challenges and concerns, both healthcare providers and technology experts expect the use of AI in healthcare to continue growing. Patients may soon encounter AI-based programs to detect emotions, which could help clinicians provide better mental and behavioral health services. Clinicians, such as neurologists and physical therapists, may use AI to measure patient movements to determine neurological and physiological impairments as well as track treatment success.

AI systems may one day independently offer diagnoses, not just support doctors in that process. Experts, however, believe many of those scenarios are years away. They stress the need for caution and governance. The healthcare industry, technology vendors creating AI-based applications and regulators need strong governance programs in place to ensure that intelligent systems are -- and can remain -- accurate and reliable.

Dig Deeper on Enterprise applications of AI

Business Analytics
Data Management