Dengke Teng1, Ping Fu2, Wenjia Li3, Feng Guo4, Hui Wang5. 1. Department of Ultrasound, China-Japan Union Hospital of Jilin University, Changchun, 130000, Jilin, China. 2. Department of Ultrasound, The First Affiliated Hospital of Nanchang University, Nanchang, 330006, Jiangxi, China. 3. Department of Breast Surgery, China-Japan Union Hospital of Jilin University, Changchun, 130000, Jilin, China. 4. Department of Ultrasound, China-Japan Union Hospital of Jilin University, Changchun, 130000, Jilin, China. guofeng_tdk@163.com. 5. Department of Ultrasound, China-Japan Union Hospital of Jilin University, Changchun, 130000, Jilin, China. wanghui_tdk@163.com.
Abstract
PURPOSE: The Thyroid Imaging, Reporting and Data System (TI-RADS) from the American College of Radiology (ACR) has been used since 2017 for the evaluation of thyroid nodules. The purpose of this study is to assess the learnability and reproducibility of TI-RADS in postgraduate freshmen. METHODS: This was a retrospective study involving 400 nodules with a final diagnosis following ultrasound (US) examination. The nodules were randomized into eight groups (50/group). Three postgraduate freshmen and three experts evaluated the nodules according to ACR TI-RADS without knowledge of the final diagnosis. After evaluating each group, training was carried out based on the inconsistencies of the freshmen/experts. Training was stopped after 200 nodules because the κ value showed almost perfect concordance. Three months later, the 50 nodules of Group 4 (the last evaluated group) were re-evaluated to assess the reproducibility of ACR TI-RADS. RESULTS: The diagnostic accuracy of the three postgraduate freshmen increased from 60%, 48%, and 46% in Group 1 to 80%, 78%, and 72% in Group 4 (P= 0.029, 0.002, and 0.008), respectively. After training, the diagnostic accuracy of the postgraduate freshmen was close to that of the experts (84%). For the US features, the postgraduate freshmen were consistent with the experts (all κ > 0.6). When re-evaluating Group 4 three months later, the five features had substantial to almost perfect agreement for the same researcher (all κ > 0.7). CONCLUSION: Based on experts' consensus, ACR TI-RADS can be learned well, and its reproducibility is excellent.
RCT Entities:
PURPOSE: The Thyroid Imaging, Reporting and Data System (TI-RADS) from the American College of Radiology (ACR) has been used since 2017 for the evaluation of thyroid nodules. The purpose of this study is to assess the learnability and reproducibility of TI-RADS in postgraduate freshmen. METHODS: This was a retrospective study involving 400 nodules with a final diagnosis following ultrasound (US) examination. The nodules were randomized into eight groups (50/group). Three postgraduate freshmen and three experts evaluated the nodules according to ACR TI-RADS without knowledge of the final diagnosis. After evaluating each group, training was carried out based on the inconsistencies of the freshmen/experts. Training was stopped after 200 nodules because the κ value showed almost perfect concordance. Three months later, the 50 nodules of Group 4 (the last evaluated group) were re-evaluated to assess the reproducibility of ACR TI-RADS. RESULTS: The diagnostic accuracy of the three postgraduate freshmen increased from 60%, 48%, and 46% in Group 1 to 80%, 78%, and 72% in Group 4 (P= 0.029, 0.002, and 0.008), respectively. After training, the diagnostic accuracy of the postgraduate freshmen was close to that of the experts (84%). For the US features, the postgraduate freshmen were consistent with the experts (all κ > 0.6). When re-evaluating Group 4 three months later, the five features had substantial to almost perfect agreement for the same researcher (all κ > 0.7). CONCLUSION: Based on experts' consensus, ACR TI-RADS can be learned well, and its reproducibility is excellent.
Authors: Franklin N Tessler; William D Middleton; Edward G Grant; Jenny K Hoang; Lincoln L Berland; Sharlene A Teefey; John J Cronan; Michael D Beland; Terry S Desser; Mary C Frates; Lynwood W Hammers; Ulrike M Hamper; Jill E Langer; Carl C Reading; Leslie M Scoutt; A Thomas Stavros Journal: J Am Coll Radiol Date: 2017-04-02 Impact factor: 5.532
Authors: Andrew S Griffin; Jason Mitsky; Upma Rawal; Abraham J Bronner; Franklin N Tessler; Jenny K Hoang Journal: J Am Coll Radiol Date: 2018-03-02 Impact factor: 5.532
Authors: R Michael Tuttle; James A Fagin; Gerald Minkowitz; Richard J Wong; Benjamin Roman; Snehal Patel; Brian Untch; Ian Ganly; Ashok R Shaha; Jatin P Shah; Mark Pace; Duan Li; Ariadne Bach; Oscar Lin; Adrian Whiting; Ronald Ghossein; Inigo Landa; Mona Sabra; Laura Boucai; Stephanie Fish; Luc G T Morris Journal: JAMA Otolaryngol Head Neck Surg Date: 2017-10-01 Impact factor: 6.223
Authors: Yang Du; Meredith Bara; Prayash Katlariwala; Roger Croutze; Katrin Resch; Jonathan Porter; Medica Sam; Mitchell P Wilson; Gavin Low Journal: World J Radiol Date: 2022-01-28