Skip to main content

ORIGINAL RESEARCH article

Front. Endocrinol.
Sec. Thyroid Endocrinology
Volume 16 - 2025 | doi: 10.3389/fendo.2025.1428888
This article is part of the Research Topic Advances in precision medicine in the management of thyroid nodules and thyroid cancer View all 36 articles

Ultrasonic Radiomics in predicting pathologic type for thyroid cancer: A preliminary study using radiomics features for predicting medullary thyroid carcinoma

Provisionally accepted
Dai Zhang Dai Zhang Fan Yang Fan Yang Wenjing Hou Wenjing Hou Ying Wang Ying Wang Jiali Mu Jiali Mu Hailing Wang Hailing Wang Xi Wei Xi Wei *
  • Department of Diagnostic and Therapeutic Ultrasonography, Tianjin Medical University Cancer Institute and Hospital, Tianjin, China,, Tianjin, China

The final, formatted version of the article will be published soon.

    Traditional ultrasound images are not sufficient to diagnose medullary thyroid cancer (MTC) accurately and effectively. We collected images from the Department of Diagnostic and Therapeutic Ultrasonography, Tianjin Medical University Cancer Institute and Hospital. Objective to establish a standard-based ultrasound imaging model for preoperative differentiation of MTC from papillary thyroid carcinoma (PTC). A retrospective cross-sectional study was conducted in a cohort of 213 patients with thyroid cancer who underwent preoperative ultrasound imaging. It included 82 patients with MTC (90 lesions) and 131 patients with PTC (135 lesions). We constructed clinical model, radiomics model and comprehensive model by executing machine learning algorithms based on baseline clinical, pathological characteristics and ultrasound image data, respectively. Our study showed that the comprehensive model observed the highest diagnostic efficacy in differentiating MTC from PTC with AUC, sensitivity, specificity, positive predictive value, negative predictive value and accuracy of 0.93, 0.88, 0.82, 0.77, 0.91, 85.8%. Delong test results showed that the comprehensive model was significantly better than the clinical model (Z=-3.791, P<0.001) and the radiomics model (Z=-2.017, P=0.044). Calibration curves indicated the comprehensive model and the radiomics model exhibited better stability than the clinical model. Decision curves analysis (DCA) demonstrated that the comprehensive model had the highest clinical net benefit. Radiomics model is effective in identifying MTC and PTC preoperatively, and the comprehensive model is better. This approach can aid in identifying the pathologic types of thyroid nodule before clinical operation, supporting personalized medicine in the decision-making process.

    Keywords: ultrasound, medullary thyroid carcinoma, Papillary thyroid carcinoma, Radiomics, personalized medicine

    Received: 07 May 2024; Accepted: 13 Jan 2025.

    Copyright: © 2025 Zhang, Yang, Hou, Wang, Mu, Wang and Wei. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

    * Correspondence: Xi Wei, Department of Diagnostic and Therapeutic Ultrasonography, Tianjin Medical University Cancer Institute and Hospital, Tianjin, China,, Tianjin, China

    Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.