Artificial Intelligence-Enabled Analysis of Radiology Reports: Epidemiology and Consequences of Incidental Thyroid Findings
🤖 AI Summary
Incidental thyroid findings (ITFs) detected during non-thyroid–indicated imaging examinations contribute significantly to thyroid cancer overdiagnosis, yet their epidemiological characteristics and mechanistic role remain poorly characterized.
Method: We developed the first Transformer-based cross-modal natural language processing pipeline to automatically extract and structure ITF features from 115,000 heterogeneous radiology reports. Integrating a retrospective cohort design with logistic regression analysis, we quantified ITF prevalence and associated downstream diagnostic and therapeutic interventions.
Contribution/Results: ITFs were identified in 7.8% of examinations, markedly increasing the odds of subsequent ultrasound evaluation, fine-needle aspiration, and surgical intervention. Among ITF-related thyroid cancer diagnoses, 83% were low-risk papillary carcinomas. These findings establish ITFs as a key upstream driver of thyroid cancer overdiagnosis, providing empirical evidence to refine risk stratification and surveillance guidelines for incidental thyroid nodules.
Application Category
📝 Abstract
Importance Incidental thyroid findings (ITFs) are increasingly detected on imaging performed for non-thyroid indications. Their prevalence, features, and clinical consequences remain undefined. Objective To develop, validate, and deploy a natural language processing (NLP) pipeline to identify ITFs in radiology reports and assess their prevalence, features, and clinical outcomes. Design, Setting, and Participants Retrospective cohort of adults without prior thyroid disease undergoing thyroid-capturing imaging at Mayo Clinic sites from July 1, 2017, to September 30, 2023. A transformer-based NLP pipeline identified ITFs and extracted nodule characteristics from image reports from multiple modalities and body regions. Main Outcomes and Measures Prevalence of ITFs, downstream thyroid ultrasound, biopsy, thyroidectomy, and thyroid cancer diagnosis. Logistic regression identified demographic and imaging-related factors. Results Among 115,683 patients (mean age, 56.8 [SD 17.2] years; 52.9% women), 9,077 (7.8%) had an ITF, of which 92.9% were nodules. ITFs were more likely in women, older adults, those with higher BMI, and when imaging was ordered by oncology or internal medicine. Compared with chest CT, ITFs were more likely via neck CT, PET, and nuclear medicine scans. Nodule characteristics were poorly documented, with size reported in 44% and other features in fewer than 15% (e.g. calcifications). Compared with patients without ITFs, those with ITFs had higher odds of thyroid nodule diagnosis, biopsy, thyroidectomy and thyroid cancer diagnosis. Most cancers were papillary, and larger when detected after ITFs vs no ITF. Conclusions ITFs were common and strongly associated with cascades leading to the detection of small, low-risk cancers. These findings underscore the role of ITFs in thyroid cancer overdiagnosis and the need for standardized reporting and more selective follow-up.
Problem
Research questions and friction points this paper is trying to address.
Developing NLP pipeline to identify incidental thyroid findings in radiology reports
Assessing prevalence and clinical outcomes of incidental thyroid findings
Investigating thyroid cancer overdiagnosis linked to incidental findings
Innovation
Methods, ideas, or system contributions that make the work stand out.
Transformer-based NLP pipeline identifies incidental thyroid findings
Natural language processing extracts nodule characteristics from reports
Retrospective analysis assesses prevalence and clinical outcomes
🔎 Similar Papers
No similar papers found.
