While generative AI systems like ChatGPT, Copilot and Gemini have captured the general public鈥檚 imagination recently, other forms of artificial intelligence (AI) and machine learning have been used across the healthcare sector for years. They are now coming, however, under increased government scrutiny. For example, in March of this year , which provides safeguards around the use of AI. On this side of the pond, in October of last year establishing an HHS AI Task Force to oversee the use of AI in healthcare.

Given the growing popularity and regulatory pressures around AI, the healthcare sector must urgently address questions about how to use AI responsibly. 

野花社区 is taking a proactive approach to ensure we use this promising technology in a way that is safe, secure, fair and transparent. To this end, we鈥檝e established a Responsible Data Use Committee and are not yet using generative AI in our applications. The committee is comprised of representatives from various departments in 野花社区 and methodically evaluates risk as it relates to our use of AI. Members of this committee share some of their key considerations to help clients evaluate AI solutions. 

If you have predictions, you have to have confidence in those predictions

With any solution that鈥檚 probabilistic in nature, you have to know how confident you can be in the solution鈥檚 predictions. Is the vendor 99% sure that the solution鈥檚 predictions are accurate? Or only 90% sure? 鈥淚f you don鈥檛 know how certain you are in a prediction, you鈥檒l be either overconfident (and that leads to you being wrong a lot) or, if there鈥檚 any uncertainty, you won鈥檛 use the prediction,鈥 says 惭耻濒迟颈笔濒补苍鈥檚 VP of Data Science, Ben Perryman. When evaluating a vendor of an AI solution, the vendor should be able to provide you with a confidence interval that expresses the degree of uncertainty around its estimates. 

Where does your liability end with the use of AI in healthcare?

The idea of supply line responsibility, or how to assign responsibility along the supply chain for AI鈥檚 use, is a critical piece of the EU AI Act. 鈥淵ou have to be pretty specific about where the customer鈥檚 liability ends and where the vendor鈥檚 liability starts,鈥 says Perryman. It鈥檚 the AI developers鈥 responsibility to develop models that solve the problems they鈥檙e intending to solve. They must also train those models on representative data. Users, for their part, must appropriately use their AI systems. Your AI vendor should help you understand exactly what the system is and isn鈥檛 capable of so that you can use it responsibly.

Getting back to basics: improving data quality鈥

Biased AI systems are a critical risk that everyone in the healthcare sector should look to mitigate. Many attempts to mitigate bias are post hoc, or after the fact. 鈥淭hat鈥檚 like buying a bullet-proof vest versus taking the bullets out of the gun,鈥 Perryman says. 鈥淵ou鈥檙e addressing the problem after it鈥檚 already a problem.鈥 Instead, Perryman advocates for a data-centric approach to mitigating bias. His team focuses on building representative data sets so that the AI models trained on them are not biased.   

Perryman believes this approach will gain steam in the industry. 鈥淵ou鈥檙e going to see a back to basics in health care to try to improve data quality,鈥 he says.   

惭耻濒迟颈笔濒补苍鈥檚 PlanOptix platform, which aggregates and enriches price transparency data, is a prime example of 惭耻濒迟颈笔濒补苍鈥檚 emphasis on data quality. To ensure only relevant, representative data is used in its calculations, the platform deploys machine-learning algorithms to eliminate any outlier data. The innovative platform recently won the 鈥淏est Overall Healthcare Data Analytics Platform鈥 award in the 2024 MedTech Breakthrough Awards program. 

Black box vs model transparency

Joshua Rice, 惭耻濒迟颈笔濒补苍鈥檚 Director of Data Science, emphasizes model transparency when it comes to evaluating AI. 鈥淵our vendor should understand how the machine learning model translates inputs into outputs. It shouldn鈥檛 be a black box, and given the current state of technology, it doesn鈥檛 have to be,鈥 Rice says. Your AI vendors should be able to provide you with impact assessments and error metrics to help you understand the accuracy, validity and reliability of its model. Your vendor should also have internal model documentation and governance protocols. 

The AI imperative鈥

Those who are concerned about the risks of using AI in healthcare should also consider the risks of not using AI, says Sean Crandell, 惭耻濒迟颈笔濒补苍鈥檚 Senior Vice President of Healthcare Economics. 鈥淭his technology is like Edison鈥檚 lightbulb. It鈥檚 that transformative,鈥 Crandell explains. 鈥淲e have an obligation to use it.鈥

惭耻濒迟颈笔濒补苍鈥檚 BenInsights solution, for example, uses advanced data analytics to perform millions of calculations in seconds. This allows it to more accurately predict a health plan鈥檚 risks and recommend actions to take to reduce that risk. Not using AI in this way and continuing to leave healthcare payors in the dark would be irresponsible, according to Crandell.  

Learn more about 惭耻濒迟颈笔濒补苍鈥檚 use of AI and how it can help your organization improve its performance.