Interpretable Data Mining of Follicular Thyroid Cancer Ultrasound Features Using Enhanced Association Rules

2025-09-08

AI Summary

Follicular thyroid carcinoma (FTC) lacks specific ultrasound features, leading to challenging preoperative diagnosis and limited supporting evidence in the literature. To address this, we propose an enhanced association rule mining method integrated with SHAP-based interpretability, applied to a cohort of 1,673 thyroid nodules with combined clinical and ultrasound data. Our approach innovatively quantifies the malignant association of imaging patterns—including “nodule-in-nodule” and trabecular architecture—with FTC, and for the first time identifies low TSH levels and coexisting Hashimoto’s thyroiditis as strong predictive biomarkers. By embedding domain-aware constraints and interpretability mechanisms, the method significantly improves both medical interpretability and clinical utility of derived rules. It identifies multiple high-confidence preoperative diagnostic markers, achieving robust statistical support (confidence ≥ 0.85, lift ≥ 3.2). These findings provide novel, evidence-based criteria for FTC risk stratification and precision diagnosis.

Technology Category

• Data Mining & Knowledge Management: Rule Mining & Pattern Mining
• Machine Learning: Feature Construction/Reformulation
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Purpose: Thyroid cancer has been a common cancer. Papillary thyroid cancer and follicular thyroid cancer are the two most common types of thyroid cancer. Follicular thyroid cancer lacks distinctive ultrasound signs and is more difficult to diagnose preoperatively than the more prevalent papillary thyroid cancer, and the clinical studies associated with it are less well established. We aimed to analyze the clinical data of follicular thyroid cancer based on a novel data mining tool to identify some clinical indications that may help in preoperative diagnosis. Methods: We performed a retrospective analysis based on case data collected by the Department of General Surgery of Peking University Third Hospital between 2010 and 2023. Unlike traditional statistical methods, we improved the association rule mining, a classical data mining method, and proposed new analytical metrics reflecting the malignant association between clinical indications and cancer with the help of the idea of SHAP method in interpretable machine learning. Results: The dataset was preprocessed to contain 1673 cases (in terms of nodes rather than patients), of which 1414 were benign and 259 were malignant nodes. Our analysis pointed out that in addition to some common indicators (e.g., irregular or lobulated nodal margins, uneven thickness halo, hypoechogenicity), there were also some indicators with strong malignant associations, such as nodule-in-nodule pattern, trabecular pattern, and low TSH scores. In addition, our results suggest that the combination of Hashimoto's thyroiditis may also have a strong malignant association. Conclusion: In the preoperative diagnosis of nodules suspected of follicular thyroid cancer, multiple clinical indications should be considered for a more accurate diagnosis. The diverse malignant associations identified in our study may serve as a reference for clinicians in related fields.

Problem

• Diagnosing follicular thyroid cancer preoperatively using ultrasound
• Identifying clinical indicators for follicular thyroid cancer malignancy
• Developing interpretable data mining methods for cancer feature analysis

Innovation

• Enhanced association rule mining method
• New analytical metrics with SHAP
• Interpretable data mining for ultrasound features