Related Experiment Video
Updated: Aug 11, 2025

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
Published on: April 21, 2023
A nomogram based on ultrasound characteristics to predict large-number cervical lymph node metastasis in papillary
Meiwu Zhang1, Yan Zhang1, Yuqin Qiu1
1Department of Interventional Therapy, Hwa Mei Hospital, University of Chinese Academy of Sciences, Zhejiang 315010, P.R. China.
Abstract:
To establish a nomogram for predicting large-number cervical lymph node metastases (LNMs) of primary papillary thyroid carcinoma (PTC) based on ultrasound characteristics. This retrospective study included patients with PTC diagnosed by pathological examination and who underwent surgery between August 2015 and May 2021 at Hwa Mei Hospital, University of Chinese Academy of Sciences (Ningbo, China). Large-number LNM was defined as >5 lymph nodes with metastases. The patients were propensity score-matched (PSM) for age and sex. A multivariable analysis was used to determine the risk factors for massive LNM. After PSM, the 78 patients with large-number LNM were matched with 312 patients with small-number LNM. Compared with the patients with small-number LNM, those with large-number LNM had larger tumors (13.0 ± 7.7 vs. 6.8 ± 3.8 mm, p < 0.001), and higher frequencies of multifocal nodules (42.3% vs. 22.4%, p < 0.001), taller-than-wide shape (82.1% vs. 56.7%, p < 0.001), calcifications (76.9% vs. 47.4%, p < 0.001), microcalcifications (68.0% vs. 36.5%, p < 0.001), capsule invasion (32.1% vs. 17.6%, p = 0.005), and ultrasound diagnosis of LNM (44.9% vs. 9.3%, p < 0.001). The multivariable analysis showed that nodule size (OR = 1.19, 95%CI: 1.11-1.27, p < 0.001), multifocal disease (OR = 2.50, 95%CI: 1.30-4.80, p = 0.006), taller-than-wide shape (OR = 0.45, 95%CI: 0.22-0.93, p = 0.032), and ultrasound diagnosis of LNM (OR = 5.57, 95%CI: 2.73-11.37, p < 0.001) were independently associated with large-number LNM. A nomogram was built, and the area under the receiver operating characteristics curve was 0.86 (95%CI: 0.81-0.90). A nomogram was successfully built to predict large-number LNM in patients with PTC, based on nodule size, multifocality, taller-than-wide shape, and ultrasound diagnosis of LNM.
More Related Videos
07:13Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model
Published on: April 18, 2025
05:52Sentinel Lymph Node Mapping and Biopsy for Endometrial Cancer at Early Stage with Laparoscopy
Published on: August 19, 2021