Across industries, artificial intelligence (AI) is transforming the way we operate. For health care providers, deploying AI in their practice can enhance patient outcomes and streamline everyday clinical operations. However, it’s also important to be aware of the concerns surrounding AI in health care, including data privacy, biases, and regulatory and ethical issues.
During the 2024 VTN Summit, Brian Gunnell, Senior System Engineer and Team Lead for the UVA Office of Telemedicine, shared insights about AI in health care, including the advantages of integrating AI technology into processes and considerations to be cautious of. Below are five takeaways from Gunnell’s presentation.
Automating Tasks and Streamlining Processes
With virtual assistants, providers can automate processes for appointment scheduling, intake of medical history, and offer personalized support for any questions that may arise. AI-powered tools can also listen in on conversations, support notetaking and scribing, and offer insights based on recordings. These automations can help simplify tasks and free up providers to focus more on their patients, improving the overall quality of patient visits.
This extra support may also help prevent burnout within the industry as processes become more efficient.
Predictive Analytics and Enhanced Image Diagnosis
AI has the ability to quickly analyze and record data, which can improve diagnosis accuracy and treatment effectiveness through predictive analytics. For example, medical images like X-rays, CT scans, and MRIs can be quickly analyzed to accurately assist in early disease detection and diagnosis. AI can even analyze a person’s genome to identify genetic predispositions to certain diseases. This allows providers more insight and proactivity when developing personalized treatment plans.
“AI can identify early detection of diseases,” Gunnell said. “It’s going to light up any red flags and tell a health care provider to dig in further.”
Mitigating Bias
With AI being programmed by humans, it often can perpetuate and amplify the biases that exist in health care. This may lead to inaccurate diagnosis and treatment and compromise the quality of care for all patients. These AI biases can stem from data that is incomplete or inaccurate, flawed programming, or even human prejudices. It’s important to regularly audit and test for biases to ensure fairness while staying up to date with the latest standards for data diversity, inclusivity, and generalizability.
Upholding Data Privacy and Security
AI systems consume vast amounts of data to improve performance and better understand objectives. “This raises concerns about the privacy and security of the data AI is consuming, how it’s stored, and who has access,” Gunnell said. “That can be pretty scary for some of the patients. Being clear with patients and making sure they understand what’s happening with their data is going to be a big part of this.”
Providers will need to be up-to-date and compliant with all HIPAA regulations. It’s crucial to ensure all patient data collected by AI is encrypted and all systems are up to date to prevent unauthorized access. Policies surrounding who owns and controls patient data and how it is shared should be clearly outlined and established. AI systems should also be designed to collect and process the minimum amount of personal data to limit exposure and privacy risks.
Cultivating Patient Trust and Acceptance
The topic of AI is still a relatively new one, leaving room for questions and concerns. Transparency will be crucial in building trust between providers and patients as AI is involved. Practices will need to create ethical internal guidelines for AI use and set boundaries for the way the technology is deployed, establishing alignment with policies and procedures. Taking the time to educate patients on the use of AI and reassuring them that a health care professional is ultimately the person making the decisions will help ease concerns and reinforce the role of AI in practice.
As AI continues to advance in health care, providers hold the responsibility of ensuring that innovative tools are used thoughtfully and ethically.
To access Gunnell’s full presentation on this topic, recordings are available for purchase here.