Integrating Artificial Intelligence in Thyroid Nodule Management: Clinical Outcomes and Cost-effectiveness Analysis
Authors' conclusion
KOIOS AI is a cost-effective alternative, particularly in reducing overdiagnosis and overtreatment for benign TNs. Prospective, real-life studies are needed to validate these findings and explore long-term implications.
Study context
- Population
- Patients who underwent thyroid surgery, May 2022–November 2024
- Participants
- 176 patients
- Comparison
- KOIOS artificial intelligence (AI) vs expert and novice operators
- Outcome
- Diagnostic performance (AUC)
Methodological detailsdeterministic and probabilistic sensitivity analyses
- Deterministic and probabilistic sensitivity analyses
Performance
As reported in the abstract
| Test or model | AUC |
|---|---|
| KOIOS AI | 0.794 (95% CI 0.718-0.871) |
| Expert operator | 0.784 (95% CI 0.706-0.861) |
| Novice operator | 0.619 (95% CI 0.526-0.711) |
AUC
KOIOS AI0.794 (95% CI 0.718-0.871)
Expert operator0.784 (95% CI 0.706-0.861)
Novice operator0.619 (95% CI 0.526-0.711)
- KOIOS AI vs Expert operator: P = 0.754 — similar diagnostic performance
- KOIOS AI vs Novice operator: P < 0.001 — KOIOS significantly outperformed
Further findings
Incremental cost-effectiveness ratio (ICER): cost per additional optimal care decision (third-party payer, 1-year horizon)
- Cost per additional optimal care decision; KOIOS dominant and cost-saving
- -€8085.56
- Deterministic sensitivity analysis: surgical costs main drivers of variability; probabilistic analysis consistently favoured KOIOS
Original publication
Integrating Artificial Intelligence in Thyroid Nodule Management: Clinical Outcomes and Cost-effectiveness Analysis. J Clin Endocrinol Metab. Published 12 July 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →