To discuss the challenges limiting the adoption of AI technologies, such as AI-enabled ultrasound and multimodal data integration, in thyroid care despite their advancements.
Key Findings:
Current AI models in thyroid care often fall short in performance and usability, impacting clinical decision-making.
Economic barriers exist due to unclear return on investment for institutions, hindering widespread adoption.
Emerging AI technologies, such as AI-enabled ultrasound and multimodal data integration, show promise but have limitations that need addressing.
Many AI tools lack prospective validation in real-world clinical settings, raising concerns about their reliability.
Interpretation:
The integration of AI in thyroid care is hindered by performance issues, economic factors, and a lack of rigorous validation, despite the potential benefits of new technologies that could enhance patient outcomes.
Limitations:
AI models may increase clinical workload rather than reduce it, complicating the workflow for healthcare providers.
Strong performance in controlled settings does not guarantee effectiveness in real-world applications, which is critical for patient safety.
Evidence gaps exist regarding the prospective validation of AI tools, which could lead to misinformed clinical decisions.
Conclusion:
For AI innovations in endocrinology to be effective, they must be validated rigorously, integrated seamlessly into workflows, demonstrate clear clinical value, and address economic barriers to adoption.
AACE Endocrine AI
is published by Conexiant under a license arrangement with the American Association of Clinical Endocrinology, Inc. (AACE®). The ideas and opinions expressed in
AACE Endocrine AI do not necessarily reflect those of Conexiant or AACE. For more information, see Policies.