Value of deep learning models based on ultrasonic dynamic videos for distinguishing thyroid nodules
ObjectiveThis study was designed to distinguish benign and malignant thyroid nodules by using deep learning(DL) models based on ultrasound dynamic videos.MethodsUltrasound dynamic videos of 1018 thyroid nodules were retrospectively collected from 657 patients in Zhejiang Cancer Hospital from January...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-01-01
|
| Series: | Frontiers in Oncology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fonc.2022.1066508/full |
| _version_ | 1828062586452574208 |
|---|---|
| author | Chen Ni Bojian Feng Jincao Yao Xueqin Zhou Jiafei Shen Di Ou Chanjuan Peng Dong Xu Dong Xu |
| author_facet | Chen Ni Bojian Feng Jincao Yao Xueqin Zhou Jiafei Shen Di Ou Chanjuan Peng Dong Xu Dong Xu |
| author_sort | Chen Ni |
| collection | DOAJ |
| description | ObjectiveThis study was designed to distinguish benign and malignant thyroid nodules by using deep learning(DL) models based on ultrasound dynamic videos.MethodsUltrasound dynamic videos of 1018 thyroid nodules were retrospectively collected from 657 patients in Zhejiang Cancer Hospital from January 2020 to December 2020 for the tests with 5 DL models.ResultsIn the internal test set, the area under the receiver operating characteristic curve (AUROC) was 0.929(95% CI: 0.888,0.970) for the best-performing model LSTM Two radiologists interpreted the dynamic video with AUROC values of 0.760 (95% CI: 0.653, 0.867) and 0.815 (95% CI: 0.778, 0.853). In the external test set, the best-performing DL model had AUROC values of 0.896(95% CI: 0.847,0.945), and two ultrasound radiologist had AUROC values of 0.754 (95% CI: 0.649,0.850) and 0.833 (95% CI: 0.797,0.869).ConclusionThis study demonstrates that the DL model based on ultrasound dynamic videos performs better than the ultrasound radiologists in distinguishing thyroid nodules. |
| first_indexed | 2024-04-10T22:30:08Z |
| format | Article |
| id | doaj.art-d05ba06a73224fd99f11916832c1ea2d |
| institution | Directory Open Access Journal |
| issn | 2234-943X |
| language | English |
| last_indexed | 2024-04-10T22:30:08Z |
| publishDate | 2023-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Oncology |
| spelling | doaj.art-d05ba06a73224fd99f11916832c1ea2d2023-01-17T06:18:16ZengFrontiers Media S.A.Frontiers in Oncology2234-943X2023-01-011210.3389/fonc.2022.10665081066508Value of deep learning models based on ultrasonic dynamic videos for distinguishing thyroid nodulesChen Ni0Bojian Feng1Jincao Yao2Xueqin Zhou3Jiafei Shen4Di Ou5Chanjuan Peng6Dong Xu7Dong Xu8The Second Clinical School of Zhejiang Chinese Medical University, Hangzhou, ChinaKey Laboratory of Head and Neck Cancer Translational Research of Zhejiang Province, Hangzhou, ChinaDepartment of Ultrasonography, The Cancer Hospital of the University of Chinese Academy of Sciences (Zhejiang Cancer Hospital), Institute of Basic Medicine and Cancer, Chinese Academy of Sciences, Hangzhou, Zhejiang, ChinaClinical Research Department, Esaote (Shenzhen) Medical Equipment Co., Ltd., Xinyilingyu Research Center, Shenzhen, ChinaCancer Hospital of the University of Chinese Academy of Sciences (Zhejiang Cancer Hospital), Hangzhou, China; Key Laboratory of Head and Neck Cancer Translational Research of Zhejiang Province, Hangzhou, ChinaCancer Hospital of the University of Chinese Academy of Sciences (Zhejiang Cancer Hospital), Hangzhou, China; Key Laboratory of Head and Neck Cancer Translational Research of Zhejiang Province, Hangzhou, ChinaCancer Hospital of the University of Chinese Academy of Sciences (Zhejiang Cancer Hospital), Hangzhou, China; Key Laboratory of Head and Neck Cancer Translational Research of Zhejiang Province, Hangzhou, ChinaKey Laboratory of Head and Neck Cancer Translational Research of Zhejiang Province, Hangzhou, ChinaDepartment of Ultrasonography, The Cancer Hospital of the University of Chinese Academy of Sciences (Zhejiang Cancer Hospital), Institute of Basic Medicine and Cancer, Chinese Academy of Sciences, Hangzhou, Zhejiang, ChinaObjectiveThis study was designed to distinguish benign and malignant thyroid nodules by using deep learning(DL) models based on ultrasound dynamic videos.MethodsUltrasound dynamic videos of 1018 thyroid nodules were retrospectively collected from 657 patients in Zhejiang Cancer Hospital from January 2020 to December 2020 for the tests with 5 DL models.ResultsIn the internal test set, the area under the receiver operating characteristic curve (AUROC) was 0.929(95% CI: 0.888,0.970) for the best-performing model LSTM Two radiologists interpreted the dynamic video with AUROC values of 0.760 (95% CI: 0.653, 0.867) and 0.815 (95% CI: 0.778, 0.853). In the external test set, the best-performing DL model had AUROC values of 0.896(95% CI: 0.847,0.945), and two ultrasound radiologist had AUROC values of 0.754 (95% CI: 0.649,0.850) and 0.833 (95% CI: 0.797,0.869).ConclusionThis study demonstrates that the DL model based on ultrasound dynamic videos performs better than the ultrasound radiologists in distinguishing thyroid nodules.https://www.frontiersin.org/articles/10.3389/fonc.2022.1066508/fullultrasoundthyroid nodulesdeep learningdistinguishingvideo |
| spellingShingle | Chen Ni Bojian Feng Jincao Yao Xueqin Zhou Jiafei Shen Di Ou Chanjuan Peng Dong Xu Dong Xu Value of deep learning models based on ultrasonic dynamic videos for distinguishing thyroid nodules Frontiers in Oncology ultrasound thyroid nodules deep learning distinguishing video |
| title | Value of deep learning models based on ultrasonic dynamic videos for distinguishing thyroid nodules |
| title_full | Value of deep learning models based on ultrasonic dynamic videos for distinguishing thyroid nodules |
| title_fullStr | Value of deep learning models based on ultrasonic dynamic videos for distinguishing thyroid nodules |
| title_full_unstemmed | Value of deep learning models based on ultrasonic dynamic videos for distinguishing thyroid nodules |
| title_short | Value of deep learning models based on ultrasonic dynamic videos for distinguishing thyroid nodules |
| title_sort | value of deep learning models based on ultrasonic dynamic videos for distinguishing thyroid nodules |
| topic | ultrasound thyroid nodules deep learning distinguishing video |
| url | https://www.frontiersin.org/articles/10.3389/fonc.2022.1066508/full |
| work_keys_str_mv | AT chenni valueofdeeplearningmodelsbasedonultrasonicdynamicvideosfordistinguishingthyroidnodules AT bojianfeng valueofdeeplearningmodelsbasedonultrasonicdynamicvideosfordistinguishingthyroidnodules AT jincaoyao valueofdeeplearningmodelsbasedonultrasonicdynamicvideosfordistinguishingthyroidnodules AT xueqinzhou valueofdeeplearningmodelsbasedonultrasonicdynamicvideosfordistinguishingthyroidnodules AT jiafeishen valueofdeeplearningmodelsbasedonultrasonicdynamicvideosfordistinguishingthyroidnodules AT diou valueofdeeplearningmodelsbasedonultrasonicdynamicvideosfordistinguishingthyroidnodules AT chanjuanpeng valueofdeeplearningmodelsbasedonultrasonicdynamicvideosfordistinguishingthyroidnodules AT dongxu valueofdeeplearningmodelsbasedonultrasonicdynamicvideosfordistinguishingthyroidnodules AT dongxu valueofdeeplearningmodelsbasedonultrasonicdynamicvideosfordistinguishingthyroidnodules |