AI is Running Rampant in Healthcare: Addressing the Challenges
The healthcare industry has entered a brave new world with the integration of artificial intelligence (AI) in various domains such as medical imaging, electronic medical records, and initial health checks. While AI algorithms have been instrumental in identifying patterns in X-rays and health records, they have also led to significant missteps in patient care.
Issues Arising from AI Integration
Despite the benefits, AI implementation has resulted in serious issues, including:
- Racial Bias: Algorithms have incorrectly deprioritized care for Black patients.
- Insurance Denials: Elderly patients have faced denials for procedures that would typically receive approval.
According to Annika Schoene, an assistant professor of public health and health sciences, the industry’s “pick and mix approach” to AI—devising rules on an ad-hoc basis rather than following comprehensive guidelines—has begun to show its limitations. She warns, “We use very little regulation when it comes to AI. I’m telling you right now, if we don’t get to grips with it, good luck.”
The Need for Universal Guidelines
Establishing a universal guide for integrating AI into hospitals is crucial. Such a guide would not only aid health care IT workers but also empower clinicians at a time when AI literacy remains low. Schoene is currently working on a project aimed at developing a universal guide for the ethical use of AI in healthcare.
This initiative will bring together computer scientists, public health researchers, ethicists, and healthcare professionals to tackle a critical question: How can tech experts be taught about ethics, and how can doctors—who are already attuned to ethical considerations—be educated about technology?
Turning Aspirational Values into Action
Cansu Canca, director of responsible AI practice at Northeastern’s Institute for Experiential AI, emphasizes the need to transform ethical goals into operational frameworks. The objective is to create a structure that allows these values to inform design, development decisions, and monitoring requirements using the technical language familiar to developers, while being rooted in the domain knowledge provided by healthcare experts.
Currently, over 1,200 AI-enabled medical devices have received approval from the U.S. Food and Drug Administration (FDA), but Schoene notes that only 8% have plans in place to monitor their use post-approval, raising potential security and health risks.
Collaborative Research for Effective Solutions
The initial phase of developing the guide involves collaboration with large healthcare systems to understand what healthcare workers need to learn about AI. The focus is on encouraging healthcare workers to proactively ask questions about AI, whether they are implementing or utilizing AI tools.
In Schoene’s words, “That’s why we’re so intentional about having workshops with people who don’t necessarily know the technical details.” The goal is to cultivate a culture of inquiry, prompting users to recognize potential red flags.
A Comprehensive Resource for Healthcare Systems
Schoene envisions the guide as a comprehensive resource that supports healthcare systems throughout their technological journeys. For instance, if a hospital is considering implementing a new AI-integrated image detection tool for breast cancer, administrators would refer to the guide to ensure it aligns with the evolving set of ethical guidelines. After installation, the same guide will help tech workers monitor important aspects, such as patient data privacy.
Adapting to an Evolving Landscape
Creating a universal AI ethics guide is akin to hitting a moving target. As technology evolves rapidly, even experts struggle to keep up, which necessitates that the guide remains a living document, adaptable over time. Schoene aims for this blueprint to equip healthcare professionals with the knowledge to push back against potential pitfalls or communicate effectively about the technology’s intended functions.
In conclusion, as AI continues to permeate healthcare, establishing clear ethical guidelines and enhancing AI literacy among healthcare workers is essential to harnessing the full potential of this transformative technology while safeguarding patient care.
