RT info:eu-repo/semantics/article
T1 Preliminary analysis of AI-based thyroid nodule evaluation in a non-subspecialist endocrinology setting
A1 Fernández Velasco, Pablo
A1 Estévez Asensio, Lucía
A1 Torres Torres, Beatriz
A1 Ortolá Buigues, Ana
A1 Gómez Hoyos, Emilia
A1 Delgado García, Esther
A1 Luis Román, Daniel Antonio de
A1 Díaz Soto, Gonzalo
K1 Thyroid nodule
K1 Artificial intelligence
K1 AI-DSS
K1 Ultrasound
K1 ACR TI-RADS
K1 ATA guidelines
K1 Risk stratification
K1 General endocrinology
K1 32 Ciencias Médicas
AB Purpose Thyroid nodules are commonly evaluated using ultrasound-based risk stratification systems, which rely on sub-jective descriptors. Artificial intelligence (AI) may improve assessment, but its effectiveness in non-subspecialist settings isunclear. This study evaluated the impact of an AI-based decision support system (AI-DSS) on thyroid nodule ultrasoundassessments by general endocrinologists (GE) without subspecialty thyroid imaging training.Methods A prospective cohort study was conducted on 80 patients undergoing thyroid ultrasound in GE outpatient clinics.Thyroid ultrasound was performed based on clinical judgment as part of routine care by GE. Images were retrospectivelyanalyzed using an AI-DSS (Koios DS), independently of clinician assessments. AI-DSS results were compared with initialGE evaluations and, when referred, with expert evaluations at a subspecialized thyroid nodule clinic (TNC). Agreement inultrasound features, risk classification by the American College of Radiology Thyroid Imaging Reporting and Data System(ACR TI-RADS) and American Thyroid Association guidelines, and referral recommendations was assessed.Results AI-DSS differed notably from GE, particularly assessing nodule composition (solid: 80%vs.36%,p < 0.01), echo-genicity (hypoechoic:52%vs.16%,p < 0.01), and echogenic foci (microcalcifications:10.7%vs.1.3%,p < 0.05). AI-DSSclassification led to a higher referral rate compared to GE (37.3%vs.30.7%, not statistically significant). Agreementbetween AI-DSS and GE in ACR TI-RADS scoring was moderate (r = 0.337;p < 0.001), but improved when comparing GEto AI-DSS and TNC subspecialist (r = 0.465;p < 0.05 and r = 0.607;p < 0.05, respectively).Conclusion In a non-subspecialist setting, non-adjunct AI-DSS use did not significantly improve risk stratification or reducehypothetical referrals. The system tended to overestimate risk, potentially leading to unnecessary procedures. Furtheroptimization is required for AI to function effectively in low-prevalence environment.
PB Springer
YR 2025
FD 2025
LK https://uvadoc.uva.es/handle/10324/75949
UL https://uvadoc.uva.es/handle/10324/75949
LA eng
NO Endocrine, 2025.
NO Producción Científica
DS UVaDOC
RD 14-jun-2025