AUTHOR=Xia Mengwen , Song Fulong , Zhao Yongfeng , Xie Yongzhi , Wen Yafei , Zhou Ping 

TITLE=Ultrasonography-based radiomics and computer-aided diagnosis in thyroid nodule management: performance comparison and clinical strategy optimization

JOURNAL=Frontiers in Endocrinology

VOLUME=Volume 14 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/endocrinology/articles/10.3389/fendo.2023.1140816

DOI=10.3389/fendo.2023.1140816

ISSN=1664-2392

ABSTRACT=Objectives: To develop ultrasonography (US) feature-based radiomics and computer-aided diagnosis (CAD) models for predicting malignancy in thyroid nodules, respectively, and evaluate their utility for thyroid nodule management. 
Methods: This prospective study included 262 thyroid nodules between January and June 2022. All nodules previously underwent standardized US image acquisition, and the nature of the nodules was confirmed by pathological results. The CAD model exploited two vertical US images of a thyroid nodule to differentiate lesions. The least absolute shrinkage and operator algorithm (LASSO) was applied to choose radiomics features with excellent predictive properties for building a radiomics model. Ultimately, the area under the receiver operating characteristic curve (AUC) and calibration curves were assessed to compare diagnostic performance among models. DeLong's test was used to analyze the difference between groups. Both models were used to revise ACR TI-RADS to provide biopsy recommendations, and their performance was compared with the original ACR.
Results: Of the 262 thyroid nodules, 157 were malignancies, and the remaining 105 were benignities. The diagnostic performance of radiomics, CAD, and ACR TI-RADS models was AUC of 0.915 (95% CI: 0.881-0.947), 0.814 (95% CI: 0.766-0.863) and 0.849 (95% CI: 0.804-0.894), respectively. DeLong's test showed a statistically significant between the AUC values of models (p < 0.05). Calibration curves showed good agreement in each model. The revised recommendations based on radiomics and CAD both improved sensitivity, accuracy, PPV, and NPV and reduced unnecessary FNA rates (33.3-9.7% vs. 33.3-16.7%).
Conclusion: Radiomics strategy and CAD system presented a well diagnostic performance for discriminating thyroid nodules and could be used to optimize ACR TI-RADS recommendation, which successfully reduces unnecessary biopsies, especially in the radiomics model.