AI Data Governance in Healthcare: What’s New and What’s Changing?

Why Does AI Make Data Governance in Healthcare More Complex?

Data governance refers to the policies and standards that ensure data is high-quality, easily accessible, secure and trustworthy. Tracking and maintaining the massive amounts of data that AI-backed technologies require has made data governance in healthcare more challenging in several key ways. 

Common challenges include:

Keeping Data Sets Updated

Healthcare data is constantly evolving, and AI training models must reflect those changes to ensure accuracy. “If you’re not updating the models daily or weekly, you’re going to miss things that are happening in the world and with your patients,” Godden says.

Removing Biases

Data may contain biases related to factors such as gender, race and socioeconomic status. Susan Laine, chief field technologist at Quest Software, says data teams must have a system in place to identify and remove those biases from the training data. “Data problems will only be amplified when fed into AI for things like diagnoses and treatment recommendations,” she warns. 

Identifying Responsibility and Accountability

If an AI-driven decision leads to an adverse outcome, is the developer, the user or the system itself responsible? “If you don’t have transparency around what’s happening with your data, then you won’t know the true source of the problem or where a fix is needed,” Laine says.

EXPLORE: Here are 13 ways AI enhances healthcare operations, patient care and treatments.

What Are the Benefits of AI and Data Governance?

A robust data governance framework ensures the AI model is receiving high-quality information, which reduces risk. “Data governance is like having a glass box around the AI,” says Laine. “It provides transparency into what’s feeding the AI model and who has touched that data.”

At the same time, AI itself can improve data management. It can be used for policy enforcement and security pattern analysis. For example, AI can monitor and verify that sensitive patient data is being accessed and handled properly. 

Chatbots can enhance the end-user experience by helping analysts more efficiently sort and interpret information from large data sets. 

In addition, machine learning tools can help healthcare organizations leverage larger influxes of data. The AI automatically processes and learns from the data it collects, which enables the system to continuously improve. 

How Can Organizations Set Realistic Expectations for AI Data Governance?

A common challenge, Godden says, is when leaders think they need to revitalize all of the organization’s data sets before they can generate value from an AI tool. Instead, he encourages them to adjust expectations and start with smaller goals: “Identify a business opportunity and focus on governing and cleaning only the data you need to solve that specific problem.” 

It’s important to clearly define the organization’s values and ensure that employees understand them. This provides necessary guidelines so that when a data anomaly occurs, employees can properly identify and fix it in accordance with the company’s expectations. “AI models are going to have biases, and corrections will come down to individuals making value calls,” Laine says. 

She adds that healthcare systems need to remember AI isn’t perfect. Human intervention is critical, especially when determining why an anomaly occurred in the data. “If I were a doctor, I would feel more reassured knowing a data governance team is behind the scenes verifying that the data makes sense,” Laine says.

Click the banner below to achieve meaningful AI transformation in healthcare with expert guidance.