A Deep Learning-Based Artificial Intelligence Model Assisting Thyroid Nodule Diagnosis and Management: Pilot Results for Evaluating Thyroid Malignancy in Pediatric Cohorts.

Journal: Thyroid : official journal of the American Thyroid Association
Published Date: Jun 2, 2025

Abstract

Artificial intelligence (AI) models have shown promise in predicting malignant thyroid nodules in adults; however, research on deep learning (DL) for pediatric cases is limited. We evaluated the applicability of a DL-based model for assessing thyroid nodules in children. We retrospectively identified two pediatric cohorts ( = 128; mean age 15.5 ± 2.4 years; 103 girls) who had thyroid nodule ultrasonography (US) with histological confirmation at two institutions. The AI-Thyroid DL model, originally trained on adult data, was tested on pediatric nodules in three scenarios axial US images, longitudinal US images, and both. We conducted a subgroup analysis based on the two pediatric cohorts and age groups (≥14 years vs. < 14 years) and compared the model's performance with radiologist interpretations using the Thyroid Imaging Reporting and Data System (TIRADS). Out of 156 nodules analyzed, 47 (30.1%) were malignant. AI-Thyroid demonstrated respective area under the receiver operating characteristic (AUROC), sensitivity, and specificity values of 0.913-0.929, 78.7-89.4%, and 79.8-91.7%, respectively. The AUROC values did not significantly differ across the image planes (all > 0.05) and between the two pediatric cohorts ( = 0.804). No significant differences were observed between age groups in terms of sensitivity and specificity (all > 0.05) while the AUROC values were higher for patients aged <14 years compared to those aged ≥14 years (all < 0.01). AI-Thyroid yielded the highest AUROC values, followed by ACR-TIRADS and K-TIRADS ( = 0.016 and < 0.001, respectively). AI-Thyroid demonstrated high performance in diagnosing pediatric thyroid cancer. Future research should focus on optimizing AI-Thyroid for pediatric use and exploring its role alongside tissue sampling in clinical practice.

Authors

  • Eun Ju Ha
    Department of Radiology, Severance Hospital, Research Institute of Radiological Science, Yonsei University College of Medicine, Seoul, Korea (G.R.K., E.-K.K., J.H.Y., H.J.M., J.Y.K.); Department of Radiology, Chung-Ang University Hospital, Chung-Ang University College of Medicine, Seoul, Korea (S.J.K.); Department of Radiology, Ajou University School of Medicine, Suwon, Korea (E.J.H.); Yonsei University College of Medicine, Seoul, Korea (J.Y.); and Department of Research Affairs, Biostatistics Collaboration Unit, Yonsei University College of Medicine, Seoul, Korea (H.S.L., J.H.H.).
  • Jeong Hoon Lee
    1 Division of Biomedical Informatics, Seoul National University Biomedical Informatics, Seoul National University College of Medicine , Seoul, Korea.
  • Natalie Mak
    Department of Radiology, Lucile Packard Children's Hospital, Stanford University, Palo Alto, California, USA.
  • Allison K Duh
    Department of Neurosurgery, Stanford University School of Medicine, Stanford, CA, USA.
  • Elizabeth Tong
    Department of Radiology, Stanford University, Stanford, California, USA.
  • Kristen W Yeom
    Department of Radiology, Stanford University, Stanford, California, United States of America.
  • Kara D Meister
    Department of Otolaryngology-Head & Neck Surgery, Children's Thyroid Clinic, Lucile Packard Children's Hospital, Stanford University, Palo Alto, California, USA.

Keywords

External Resources

Popular Topics
Recent Journals