Artificial intelligence-enabled analysis of radiology reports: epidemiology and outcomes of incidental thyroid findings
Authors' conclusion
ITFs were common and strongly associated with cascades leading to the detection of small, low-risk cancers, highlighting their role in overdiagnosis and the need for standardized reporting and more selective follow-up.
Study context
- Population
- Adults without prior thyroid disease undergoing thyroid-capturing imaging, July 1, 2017–September 30, 2023, Mayo Clinic sites; mean age 56.8 (SD 17.2), 52.9% women
- Participants
- 115,683 adults
- Exposure
- Patients with incidental thyroid findings (ITFs) vs without ITFs
- ITFn = 9,077
- Outcome
- ITF prevalence
Methodological detailsnatural language processing pipeline, multicentre
- Natural language processing pipeline
- Multicentre
Main finding
As reported in the abstract
| Group | Value | Nodular |
|---|---|---|
| ITF | 7.8% | 92.9% |
ITF
Value7.8%
Nodular92.9%
Observational data: associations, not effects.
Further findings
| Outcome | OR (95% CI) | |
|---|---|---|
| Thyroid nodule diagnosis | 45 (41.1–49.3) | |
| Biopsy | 46.8 (39.0–56.2) | |
| Thyroidectomy | 55.8 (31.3–99.3) | |
| Thyroid cancer diagnosis | 61.7 (38.6–98.5) | |
| 0.050.21520Odds ratio (log scale) |
Cancer size
- Cancer size — Most cancers were papillary
- 88.5%
Factors associated with ITFs
- ITFs more likely in women, older adults, higher BMI, and imaging ordered by specialties other than Emergency Medicine
- Vs chest CT, ITFs more likely via neck CT, PET and nuclear medicine scans
Documentation of nodule characteristics
- Size reported
- 44%
- Other features reported
- fewer than 15%
Original publication
Artificial intelligence-enabled analysis of radiology reports: epidemiology and outcomes of incidental thyroid findings. J Clin Endocrinol Metab. Published 27 April 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →