Clinical Report: AI Advances in Thyroid Care Face Barriers to Adoption

Overview

Revise to specify which AI models are underperforming and how they impact clinician workload.

Background

The integration of AI in thyroid care holds significant potential for improving diagnostic accuracy and patient management. However, the gap between technological advancements and clinical implementation remains a critical issue. Addressing these barriers is essential for realizing the benefits of AI in enhancing patient outcomes in endocrinology.

Data Highlights

No numerical data available in the source material.

Key Findings

• Current AI models in thyroid care often lack performance and usability, leading to skepticism among clinicians.
• Economic barriers, including unclear return on investment, impede institutional adoption of AI technologies.
• AI-enabled ultrasound platforms can enhance nodule characterization and provide automated malignancy risk scores.
• Multimodal data integration using AI can improve diagnostic performance, particularly in indeterminate cases.
• Significant evidence gaps exist, with many AI tools lacking prospective validation in real-world settings.

Clinical Implications

Clinicians should remain cautious about adopting AI tools without robust validation and clear clinical value. As reimbursement policies evolve, there may be opportunities for improved integration of AI in thyroid care, but ongoing evaluation of these technologies is necessary.

Conclusion

While AI has the potential to transform thyroid care, addressing the barriers to its adoption is crucial for ensuring its effective integration into clinical practice.

References

1. AACE Endocrine AI, AACE 2026: AI in thyroid cancer care: Progress and gaps
2. AACE Endocrine AI, AACE 2026: Will AI replace endocrinologists?
3. AACE Endocrine AI, AACE Endocrine AI is here: Why you need this now
4. 2025 American Thyroid Association Management Guidelines for Adult Patients with Differentiated Thyroid Cancer
5. Frontiers | Diagnostic performance of ultrasound characteristics-based artificial intelligence models for thyroid nodules: a systematic review and meta-analysis
6. The ASCO Post — AI Model May Aid in Screening, Staging, and Treatment Planning for Thyroid Cancer
7. Artificial intelligence-based multi-modal multi-tasks analysis of thyroid ultrasound image features predicts thyroid cancer: a multicenter study
8. 2025 American Thyroid Association Management Guidelines for Adult Patients with Differentiated Thyroid Cancer - Matthew D. Ringel, Julie Ann Sosa, Zubair Baloch, Lindsay Bischoff, Gary Bloom, Gregory A. Brent, Pamela L. Brock, Roger Chou, Robert R. Flavell, Whitney Goldner, Elizabeth G. Grubbs, Megan Haymart, Steven M. Larson, Angela M. Leung, Joseph Osborne, John A. Ridge, Bruce Robinson, David L. Steward, Ralph P. Tufano, Lori J. Wirth, 2025
9. Frontiers | Diagnostic performance of ultrasound characteristics-based artificial intelligence models for thyroid nodules: a systematic review and meta-analysis
10. From Guidelines to Intelligence: How AI Refines Thyroid Nodule Biopsy Decisions - ScienceDirect
11. Frontiers | Beyond genomics: artificial intelligence-powered diagnostics for indeterminate thyroid nodules—a systematic review and meta-analysis
12. Transcript for CDRH Webinar Final Guidance: Marketing Submission Recommendations for a Predetermined Change Control Plan for Artificial Intelligence-Enabled Device Software Functions
13. Interoperability Helps Radiology AI Deliver Value | RSNA
14. Reimbursement for Artificial Intelligence Software as a Medical Device in Radiology - ScienceDirect