AUTHOR=Jiang Shan , Xie Qingji , Li Nan , Chen Haizhen , Chen Xi 

TITLE=Modified Models for Predicting Malignancy Using Ultrasound Characters Have High Accuracy in Thyroid Nodules With Small Size

JOURNAL=Frontiers in Molecular Biosciences

VOLUME=Volume 8 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/molecular-biosciences/articles/10.3389/fmolb.2021.752417

DOI=10.3389/fmolb.2021.752417

ISSN=2296-889X

ABSTRACT=To assess the malignancy risk of thyroid nodules, five ultrasound characteristics are suggested as key diagnostic markers. European Thyroid Association Guidelines  (EU-TIRADS) and American Thyroid Association Management Guidelines 2015 (2015ATA) and are mainly used for ultrasound malignancy risk stratification. But both are less accuracy and not well classify the high risk patients in clinical examination. So,  fine needle aspiration biopsy (FNAB) were considered conclusive method. But, small nodules (< 0.5 cm) reduce the accuracy of FNAB. Thus, This study aimed to determine the diagnostic accuracy and establish a models using all ultrasound features including nodule size for predicting the malignancy of thyroid nodules in China. We applied logistic regression to the data of 1026 patients who received medical treatment at the Shanghai and Fujian. Ultrasound features including hypo-echogenicity and invasion of the thyroid capsule showed high odd ratios (OR 34.5 and 20.13) in this datasets. Four Ultrasound features revealed distinct OR in nodule groups with different size. In this study, we constructed a logistic models with accuracy up to 80% whatever the nodule size (< or > 0.5 cm) which is higher than EU-TIRADS. In addition, this model revealed that nodules with more ultrasound features has high probability of malignancy. Hence, this models could well predict malignancy in small nodules and classify the high risk patients.