Back to Liminal Blog

Securely Enabling Generative AI Use Cases in Healthcare

Explore how generative AI is being used in healthcare today, and how organizations can leverage this technology while maintaining patient safety and privacy.

Share On:
Share on LinkedIn
Share on Twitter

Generative AI is already profoundly positively impacting the healthcare industry. From reducing the administrative burden on clinicians and staff, to expediting patient research and diagnoses, to helping some of the brightest medical minds solve the hardest problems in the field, this groundbreaking technology is enabling providers to spend more time focusing on what matters most - enriching the lives of their patients. 

As healthcare organizations look to fully embrace the benefits of generative AI, they must also tackle the very real risks related to data privacy, security, and sovereignty that come with its usage. This article explores some of the current applications of generative AI in the healthcare sector, and how providers can leverage these tools while ensuring patient safety and privacy remain at the forefront of care.

Utilizing Generative AI in Healthcare

Here are several scenarios in which generative AI is being leveraged by healthcare organizations today. 

Automating Practitioner Notes:

Documenting the outcomes of a patient visit is crucial for maintaining a comprehensive understanding of the patient's care plan. However, this process can be extremely time-consuming. In an effort to streamline administrative tasks and allow physicians to dedicate more time to patient care, The University of Kansas Health System is testing a generative AI copilot that can summarize conversations between providers and patients, and produce real-time clinical documentation. With the introduction of this copilot, KU Health's care team aims to substantially cut down on the two-plus hours their physicians dedicate to patient documentation every day.

Medical Imaging and Diagnosis:

Generative AI has shown tremendous potential in the field of medical imaging and diagnosis. Advanced algorithms can analyze complex medical images, such as MRI and CT scans, with a level of precision that surpasses traditional methods. At Cleveland Clinic, for example, specialists are integrating generative AI into their stroke response protocols by sending images from a patient’s CT scan through an algorithm that can quickly identify blockages in large arteries. If occlusions are detected, the system then notifies the relevant physicians so they can determine next steps.

Clinical Decision Support:

The care transition process plays a crucial role in effective patient treatment. Knowing when and where to refer a patient for specialist follow-up is a key aspect of the care cycle, but it can be tedious. The complexities of completing the appropriate workup, selecting the right specialist, and determining whether a patient truly needs specialist care contribute to mistaken referrals, causing frustration for both patients and providers. To better equip its primary care physicians with timely and accurate knowledge, Providence Health Systems integrated generative AI into its education and referral platform to provide advice and guidance around specialty care. Initial feedback has been overwhelmingly positive, with 72% of physicians reporting improvement in their workups, 20% indicating a change in the specialists they referred to, and another 20% noting the ability to manage patient care without the need for a specialist.

Proactive Identification:

Quickly identifying medical conditions is paramount for early intervention and enabling timely and effective treatment that can significantly improve patient outcomes. Early detection not only enhances the chances of successful recovery but also reduces the potential for complications. AdventHealth is piloting generative AI-enabled tools to predict sepsis, the leading cause of fatalities in US hospitals. These tools monitor patient vitals across dozens of variables simultaneously, allowing providers to detect sepsis risk earlier and respond more quickly before the patient’s condition deteriorates. Early results from the utilization of these tools indicate an improvement in sepsis detection of 118%.

Protecting Patient Data is Critical

The applications of generative AI in healthcare are diverse and immensely impactful - and will only continue to grow. As more and more generative AI-enabled tools become available, it’s crucial to develop a strategy that addresses the data security concerns inherent with this technology to ensure PHI, PII, Intellectual Property, and other sensitive information are protected from being shared outside your organization. A layered approach that leverages policy, process, and technology can help keep patient information secure and protect against violations of data protection regulations like HIPAA.

Liminal is the security technology layer for healthcare organizations looking to take advantage of generative AI. To see the Liminal Platform in action, click here to schedule a demo.