Artificial intelligence (AI) has transformed the healthcare sector. With broadly useful applications across administrative and clinical work, healthcare professionals depend on AI in many contexts. While there are pros and cons to AI in healthcare, it’s clear that the benefits far outweigh any potential drawbacks.
AI in healthcare case studies and peer-reviewed studies offer powerful examples of how real individuals and organizations have realized positive change through the power of this technology. Let’s review four examples of artificial intelligence in healthcare.
1. AI Technology Offers Valuable Clinical Decision Support
Big data analytics and AI have transformed the way clinicians work at TidalHealth Peninsula Regional, a hospital in Maryland.
The facility identified issues with clinicians having to spend too much time searching for information. The hospital partnered with IBM to implement a clinical decision support software, IBM Micromedex with Watson, now called DynaMedex. This solution combines AI and Natural Language Processing (NLP) and AI with patients’ electronic medical records, making it easier to find relevant and useful information and provide clinical decision support.
For TidalHealth, that meant cutting down on the time providers spend on clinical searches. The many steps involved in that process, and the dozens of instances in which it may have to be completed in a single shift, took significant time away from providers.
With a dependable AI tool supporting more efficient information gathering, the hospital cut time per clinical search from 3-4 minutes to less than 1 minute. That’s more time medical professionals can instead spend with patients each day, thanks to AI optimizing a time-consuming process.
2. A Broad Goal: Leveraging Healthcare Data More Effectively
The Healthcare Information and Management Systems Society (HIMSS) details how the Mayo Clinic and Google Cloud developed an AI and ML platform to support patient care and research.
By collaborating to build a strong foundation, the provider and tech company delivered a variety of benefits to practitioners and clinicians, arming them with a wide variety of AI tools to support patient care and research. In-depth calculations, like those used to assess changes related to polycystic kidney disease, can be completed automatically. Another use ties existing EHR Data and algorithms to help assess breast cancer risk.
3. Peer-Reviewed Evidence of Real Outcomes from AI in Patient Status Review & Length of Stay Predictions
Independent research teams have taken a closer look at how AI supports clinicians and utilization review teams. Recent peer-reviewed studies examined the impact of AI-powered tools like Xsolis’ Care Level Score™ (CLS™) and predictive models at three health systems: Baylor Scott & White, Yale New Haven Health, and Mayo Clinic Health System.
Researchers observed that utilizing AI-driven solutions as validated guides in analyzing medical necessity helped bring objectivity to complex cases. Teams at these healthcare organizations reported feeling better equipped to make timely, appropriate patient status decisions. The research also documented measurable improvements in workflow efficiency and coordination between utilization review teams and clinical staff.
Together, these independent evaluations reinforce how well-implemented AI can function as a clinical and operational partner instead of an added burden or task, while allowing providers more time to focus on patient care.
4. AI Driving Efficiencies in Patient Case Review
Valley Medical Center in Renton, WA, implemented Xsolis’ Dragonfly solution, to right-size its observation rates. With AI leading to efficiencies in case review and management, the facility’s nurses could focus on clinical merit in case determinations. Instead of relying on inefficient, criteria-based solutions, they could use their skills and experience to better support patients.
In addition to right-sizing its observation rates to keep them more within the Centers for Medicare and Medicaid Services (CMS) and other local facilities’ averages, Valley Medical Center also reduced its extended stay observation rates (those patients who are discharged in an observation status who stay longer than two midnights), while also dramatically improving case review volume. The facility went from completing 60% of reviews to 100% — a 67% improvement.
The history of AI in healthcare is still being written, as AI, predictive models, algorithms and similar tools continue to transform the healthcare industry.