Image reporting and characterization system for ultrasound features of thyroid nodules: Multicentric Korean retrospective study

Jin Young Kwak, Inkyung Jung, Jung Hwan Baek, Seon Mi Baek, Nami Choi, Yoon Jung Choi, So Lyung Jung, Eun Kyung Kim, Jeong Ah Kim, Ji Hoon Kim, Kyu Sun Kim, Jeong Hyun Lee, Joon Hyung Lee, Hee Jung Moon, Won Jin Moon, Jeong Seon Park, Ji Hwa Ryu, Jung Hee Shin, Eun Ju Son, Jin Yong SungDong Gyu Na

Research output: Contribution to journalArticle

89 Citations (Scopus)

Abstract

Objective: The objective of this retrospective study was to develop and validate a simple diagnostic prediction model by using ultrasound (US) features of thyroid nodules obtained from multicenter retrospective data. Materials and Methods: Patient data were collected from 20 different institutions and the data included 2000 thyroid nodules from 1796 patients. For developing a diagnostic prediction model to estimate the malignant risk of thyroid nodules using suspicious malignant US features, we developed a training model in a subset of 1402 nodules from 1260 patients. Several suspicious malignant US features were evaluated to create the prediction model using a scoring tool. The scores for such US features were estimated by calculating odds ratios, and the risk score of malignancy for each thyroid nodule was defined as the sum of these individual scores. Later, we verified the usefulness of developed scoring system by applying into the remaining 598 nodules from 536 patients. Results: Among 2000 tumors, 1268 were benign and 732 were malignant. In our multiple regression analysis models, the following US features were statistically significant for malignant nodules when using the training data set: hypoechogenicity, marked hypoechogenicity, non-parallel orientation, microlobulated or spiculated margin, ill-defined margins, and microcalcifications. The malignancy rate was 7.3% in thyroid nodules that did not have suspicious-malignant features on US. Area under the receiver operating characteristic (ROC) curve was 0.867, which shows that the US risk score help predict thyroid malignancy well. In the test data set, the malignancy rates were 6.2% in thyroid nodules without malignant features on US. Area under the ROC curve of the test set was 0.872 when using the prediction model. Conclusion: The predictor model using suspicious malignant US features may be helpful in risk stratification of thyroid nodules.

Original languageEnglish
Pages (from-to)110-117
Number of pages8
JournalKorean journal of radiology
Volume14
Issue number1
DOIs
Publication statusPublished - 2013 Jan 1

Fingerprint

Thyroid Nodule
Retrospective Studies
Neoplasms
ROC Curve
Calcinosis
Thyroid Gland
Odds Ratio
Regression Analysis

All Science Journal Classification (ASJC) codes

  • Radiology Nuclear Medicine and imaging

Cite this

Kwak, Jin Young ; Jung, Inkyung ; Baek, Jung Hwan ; Baek, Seon Mi ; Choi, Nami ; Choi, Yoon Jung ; Jung, So Lyung ; Kim, Eun Kyung ; Kim, Jeong Ah ; Kim, Ji Hoon ; Kim, Kyu Sun ; Lee, Jeong Hyun ; Lee, Joon Hyung ; Moon, Hee Jung ; Moon, Won Jin ; Park, Jeong Seon ; Ryu, Ji Hwa ; Shin, Jung Hee ; Son, Eun Ju ; Sung, Jin Yong ; Na, Dong Gyu. / Image reporting and characterization system for ultrasound features of thyroid nodules : Multicentric Korean retrospective study. In: Korean journal of radiology. 2013 ; Vol. 14, No. 1. pp. 110-117.
@article{2518fcd82ce54e41853d6ca556e36377,
title = "Image reporting and characterization system for ultrasound features of thyroid nodules: Multicentric Korean retrospective study",
abstract = "Objective: The objective of this retrospective study was to develop and validate a simple diagnostic prediction model by using ultrasound (US) features of thyroid nodules obtained from multicenter retrospective data. Materials and Methods: Patient data were collected from 20 different institutions and the data included 2000 thyroid nodules from 1796 patients. For developing a diagnostic prediction model to estimate the malignant risk of thyroid nodules using suspicious malignant US features, we developed a training model in a subset of 1402 nodules from 1260 patients. Several suspicious malignant US features were evaluated to create the prediction model using a scoring tool. The scores for such US features were estimated by calculating odds ratios, and the risk score of malignancy for each thyroid nodule was defined as the sum of these individual scores. Later, we verified the usefulness of developed scoring system by applying into the remaining 598 nodules from 536 patients. Results: Among 2000 tumors, 1268 were benign and 732 were malignant. In our multiple regression analysis models, the following US features were statistically significant for malignant nodules when using the training data set: hypoechogenicity, marked hypoechogenicity, non-parallel orientation, microlobulated or spiculated margin, ill-defined margins, and microcalcifications. The malignancy rate was 7.3{\%} in thyroid nodules that did not have suspicious-malignant features on US. Area under the receiver operating characteristic (ROC) curve was 0.867, which shows that the US risk score help predict thyroid malignancy well. In the test data set, the malignancy rates were 6.2{\%} in thyroid nodules without malignant features on US. Area under the ROC curve of the test set was 0.872 when using the prediction model. Conclusion: The predictor model using suspicious malignant US features may be helpful in risk stratification of thyroid nodules.",
author = "Kwak, {Jin Young} and Inkyung Jung and Baek, {Jung Hwan} and Baek, {Seon Mi} and Nami Choi and Choi, {Yoon Jung} and Jung, {So Lyung} and Kim, {Eun Kyung} and Kim, {Jeong Ah} and Kim, {Ji Hoon} and Kim, {Kyu Sun} and Lee, {Jeong Hyun} and Lee, {Joon Hyung} and Moon, {Hee Jung} and Moon, {Won Jin} and Park, {Jeong Seon} and Ryu, {Ji Hwa} and Shin, {Jung Hee} and Son, {Eun Ju} and Sung, {Jin Yong} and Na, {Dong Gyu}",
year = "2013",
month = "1",
day = "1",
doi = "10.3348/kjr.2013.14.1.110",
language = "English",
volume = "14",
pages = "110--117",
journal = "Korean Journal of Radiology",
issn = "1229-6929",
publisher = "Korean Radiological Society",
number = "1",

}

Kwak, JY, Jung, I, Baek, JH, Baek, SM, Choi, N, Choi, YJ, Jung, SL, Kim, EK, Kim, JA, Kim, JH, Kim, KS, Lee, JH, Lee, JH, Moon, HJ, Moon, WJ, Park, JS, Ryu, JH, Shin, JH, Son, EJ, Sung, JY & Na, DG 2013, 'Image reporting and characterization system for ultrasound features of thyroid nodules: Multicentric Korean retrospective study', Korean journal of radiology, vol. 14, no. 1, pp. 110-117. https://doi.org/10.3348/kjr.2013.14.1.110

Image reporting and characterization system for ultrasound features of thyroid nodules : Multicentric Korean retrospective study. / Kwak, Jin Young; Jung, Inkyung; Baek, Jung Hwan; Baek, Seon Mi; Choi, Nami; Choi, Yoon Jung; Jung, So Lyung; Kim, Eun Kyung; Kim, Jeong Ah; Kim, Ji Hoon; Kim, Kyu Sun; Lee, Jeong Hyun; Lee, Joon Hyung; Moon, Hee Jung; Moon, Won Jin; Park, Jeong Seon; Ryu, Ji Hwa; Shin, Jung Hee; Son, Eun Ju; Sung, Jin Yong; Na, Dong Gyu.

In: Korean journal of radiology, Vol. 14, No. 1, 01.01.2013, p. 110-117.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Image reporting and characterization system for ultrasound features of thyroid nodules

T2 - Multicentric Korean retrospective study

AU - Kwak, Jin Young

AU - Jung, Inkyung

AU - Baek, Jung Hwan

AU - Baek, Seon Mi

AU - Choi, Nami

AU - Choi, Yoon Jung

AU - Jung, So Lyung

AU - Kim, Eun Kyung

AU - Kim, Jeong Ah

AU - Kim, Ji Hoon

AU - Kim, Kyu Sun

AU - Lee, Jeong Hyun

AU - Lee, Joon Hyung

AU - Moon, Hee Jung

AU - Moon, Won Jin

AU - Park, Jeong Seon

AU - Ryu, Ji Hwa

AU - Shin, Jung Hee

AU - Son, Eun Ju

AU - Sung, Jin Yong

AU - Na, Dong Gyu

PY - 2013/1/1

Y1 - 2013/1/1

N2 - Objective: The objective of this retrospective study was to develop and validate a simple diagnostic prediction model by using ultrasound (US) features of thyroid nodules obtained from multicenter retrospective data. Materials and Methods: Patient data were collected from 20 different institutions and the data included 2000 thyroid nodules from 1796 patients. For developing a diagnostic prediction model to estimate the malignant risk of thyroid nodules using suspicious malignant US features, we developed a training model in a subset of 1402 nodules from 1260 patients. Several suspicious malignant US features were evaluated to create the prediction model using a scoring tool. The scores for such US features were estimated by calculating odds ratios, and the risk score of malignancy for each thyroid nodule was defined as the sum of these individual scores. Later, we verified the usefulness of developed scoring system by applying into the remaining 598 nodules from 536 patients. Results: Among 2000 tumors, 1268 were benign and 732 were malignant. In our multiple regression analysis models, the following US features were statistically significant for malignant nodules when using the training data set: hypoechogenicity, marked hypoechogenicity, non-parallel orientation, microlobulated or spiculated margin, ill-defined margins, and microcalcifications. The malignancy rate was 7.3% in thyroid nodules that did not have suspicious-malignant features on US. Area under the receiver operating characteristic (ROC) curve was 0.867, which shows that the US risk score help predict thyroid malignancy well. In the test data set, the malignancy rates were 6.2% in thyroid nodules without malignant features on US. Area under the ROC curve of the test set was 0.872 when using the prediction model. Conclusion: The predictor model using suspicious malignant US features may be helpful in risk stratification of thyroid nodules.

AB - Objective: The objective of this retrospective study was to develop and validate a simple diagnostic prediction model by using ultrasound (US) features of thyroid nodules obtained from multicenter retrospective data. Materials and Methods: Patient data were collected from 20 different institutions and the data included 2000 thyroid nodules from 1796 patients. For developing a diagnostic prediction model to estimate the malignant risk of thyroid nodules using suspicious malignant US features, we developed a training model in a subset of 1402 nodules from 1260 patients. Several suspicious malignant US features were evaluated to create the prediction model using a scoring tool. The scores for such US features were estimated by calculating odds ratios, and the risk score of malignancy for each thyroid nodule was defined as the sum of these individual scores. Later, we verified the usefulness of developed scoring system by applying into the remaining 598 nodules from 536 patients. Results: Among 2000 tumors, 1268 were benign and 732 were malignant. In our multiple regression analysis models, the following US features were statistically significant for malignant nodules when using the training data set: hypoechogenicity, marked hypoechogenicity, non-parallel orientation, microlobulated or spiculated margin, ill-defined margins, and microcalcifications. The malignancy rate was 7.3% in thyroid nodules that did not have suspicious-malignant features on US. Area under the receiver operating characteristic (ROC) curve was 0.867, which shows that the US risk score help predict thyroid malignancy well. In the test data set, the malignancy rates were 6.2% in thyroid nodules without malignant features on US. Area under the ROC curve of the test set was 0.872 when using the prediction model. Conclusion: The predictor model using suspicious malignant US features may be helpful in risk stratification of thyroid nodules.

UR - http://www.scopus.com/inward/record.url?scp=84872173775&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84872173775&partnerID=8YFLogxK

U2 - 10.3348/kjr.2013.14.1.110

DO - 10.3348/kjr.2013.14.1.110

M3 - Article

C2 - 23323040

AN - SCOPUS:84872173775

VL - 14

SP - 110

EP - 117

JO - Korean Journal of Radiology

JF - Korean Journal of Radiology

SN - 1229-6929

IS - 1

ER -