Prediction of Seronegative Hashimoto's thyroiditis using machine learning models based on ultrasound radiomics: a multicenter study
Abstract Background Seronegative Hashimoto's thyroiditis is often underdiagnosed due to the lack of antibody markers. Combining ultrasound radiomics with machine learning offers potential for early detection in patients with normal thyroid function. Methods Data from 164 patients with single th...
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-04-01
|
| Series: | BMC Immunology |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12865-025-00708-5 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850183877371363328 |
|---|---|
| author | Wenjun Wu Shengsheng Yao Daming Liu Yuan Luo Yihan Sun Ting Ruan Mengyou Liu Li Shi Chang Liu Mingming Xiao Qi Zhang Zhengshuai Liu Xingai Ju Jiahao Wang Xiang Fei Li Lu Yang Gao Ying Zhang Liying Gong Xuanyu Chen Wanli Zheng Xiali Niu Xiao Yang Huimei Cao Shijie Chang Jianchun Cui Zuoxin Ma |
| author_facet | Wenjun Wu Shengsheng Yao Daming Liu Yuan Luo Yihan Sun Ting Ruan Mengyou Liu Li Shi Chang Liu Mingming Xiao Qi Zhang Zhengshuai Liu Xingai Ju Jiahao Wang Xiang Fei Li Lu Yang Gao Ying Zhang Liying Gong Xuanyu Chen Wanli Zheng Xiali Niu Xiao Yang Huimei Cao Shijie Chang Jianchun Cui Zuoxin Ma |
| author_sort | Wenjun Wu |
| collection | DOAJ |
| description | Abstract Background Seronegative Hashimoto's thyroiditis is often underdiagnosed due to the lack of antibody markers. Combining ultrasound radiomics with machine learning offers potential for early detection in patients with normal thyroid function. Methods Data from 164 patients with single thyroid lesions and normal thyroid function, treated surgically between 2016 and 2024, were retrospectively collected from four hospitals. Radiomics features were extracted from ultrasound images of non-tumorous hypoechoic areas. Pathological lymphocytic infiltration and hypoechoic ratios were evaluated by senior pathologists and ultrasound physicians. A machine learning model, CCH-NET, was developed using a random forest classifier after feature selection with Least Absolute Shrinkage and Selection Operator (LASSO) regression. The model was trained and tested with an 80:20 split and compared to senior ultrasound physicians. Results The CCH-NET model achieved a sensitivity of 0.762, specificity of 0.714, and an area under the curve (AUC) of 0.8248, outperforming senior ultrasound physicians (AUC = 0.681). It maintained consistent accuracy across test sets, with F1 scores of 0.778 and 0.720 in Test_1 and Test_2, respectively, and exhibited superior predictive rates. Conclusion The CCH-NET model enhances accuracy in detecting early Seronegative Hashimoto's thyroiditis over senior ultrasound physicians. Ethics No. [2023] H013 Trial registration Chinese Clinical Trial Registry;CTR2400092179; 12 November 2024. |
| format | Article |
| id | doaj-art-a73cf36f85514af698b92b9736e1e95c |
| institution | OA Journals |
| issn | 1471-2172 |
| language | English |
| publishDate | 2025-04-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Immunology |
| spelling | doaj-art-a73cf36f85514af698b92b9736e1e95c2025-08-20T02:17:13ZengBMCBMC Immunology1471-21722025-04-0126111210.1186/s12865-025-00708-5Prediction of Seronegative Hashimoto's thyroiditis using machine learning models based on ultrasound radiomics: a multicenter studyWenjun Wu0Shengsheng Yao1Daming Liu2Yuan Luo3Yihan Sun4Ting Ruan5Mengyou Liu6Li Shi7Chang Liu8Mingming Xiao9Qi Zhang10Zhengshuai Liu11Xingai Ju12Jiahao Wang13Xiang Fei14Li Lu15Yang Gao16Ying Zhang17Liying Gong18Xuanyu Chen19Wanli Zheng20Xiali Niu21Xiao Yang22Huimei Cao23Shijie Chang24Jianchun Cui25Zuoxin Ma26Liaoning University of Chinese MedicineDepartment of Thyroid and Breast Surgery, People’s Hospital of China Medical University (Liaoning Provincial People’s Hospital)Department of Ultrasound Medicine, People’s Hospital of China Medical University (Liaoning Provincial People’s Hospital)Department of Pathology, People’s Hospital of China Medical University (Liaoning Provincial People’s Hospital)Department of Cardiology, People’s Hospital of China Medical University (Liaoning Provincial People’s Hospital)Liaoning University of Chinese MedicineDepartment of Thyroid Surgery, Lixin County People’s HospitalJinzhou Medical College Linghai Dalinghe HospitalSchool of Intelligent Medicine, China Medical UniversityDepartment of Pathology, People’s Hospital of China Medical University (Liaoning Provincial People’s Hospital)Department of Ultrasound Medicine, People’s Hospital of China Medical University (Liaoning Provincial People’s Hospital)Fengcheng Phoenix HospitalDepartment of General Medicine, People’s Hospital of China Medical University (Liaoning Provincial People’s Hospital)School of Intelligent Medicine, China Medical UniversityDepartment of Thyroid and Breast Surgery, People’s Hospital of China Medical University (Liaoning Provincial People’s Hospital)Department of Endocrinology, People’s Hospital of China Medical University (Liaoning Provincial People’s Hospital)Department of Thyroid and Breast Surgery, People’s Hospital of China Medical University (Liaoning Provincial People’s Hospital)Department of Thyroid and Breast Surgery, People’s Hospital of China Medical University (Liaoning Provincial People’s Hospital)Department of Medical Laboratory, People’s Hospital of China Medical University (Liaoning Provincial People’s Hospital)Jinzhou Medical UniversityJinzhou Medical UniversityDepartment of General Medicine, People’s Hospital of China Medical University (Liaoning Provincial People’s Hospital)Department of Pathology, People’s Hospital of China Medical University (Liaoning Provincial People’s Hospital)Jinzhou Medical College Linghai Dalinghe HospitalSchool of Intelligent Medicine, China Medical UniversityDepartment of Thyroid and Breast Surgery, People’s Hospital of China Medical University (Liaoning Provincial People’s Hospital)Department of Medical Laboratory, People’s Hospital of China Medical University (Liaoning Provincial People’s Hospital)Abstract Background Seronegative Hashimoto's thyroiditis is often underdiagnosed due to the lack of antibody markers. Combining ultrasound radiomics with machine learning offers potential for early detection in patients with normal thyroid function. Methods Data from 164 patients with single thyroid lesions and normal thyroid function, treated surgically between 2016 and 2024, were retrospectively collected from four hospitals. Radiomics features were extracted from ultrasound images of non-tumorous hypoechoic areas. Pathological lymphocytic infiltration and hypoechoic ratios were evaluated by senior pathologists and ultrasound physicians. A machine learning model, CCH-NET, was developed using a random forest classifier after feature selection with Least Absolute Shrinkage and Selection Operator (LASSO) regression. The model was trained and tested with an 80:20 split and compared to senior ultrasound physicians. Results The CCH-NET model achieved a sensitivity of 0.762, specificity of 0.714, and an area under the curve (AUC) of 0.8248, outperforming senior ultrasound physicians (AUC = 0.681). It maintained consistent accuracy across test sets, with F1 scores of 0.778 and 0.720 in Test_1 and Test_2, respectively, and exhibited superior predictive rates. Conclusion The CCH-NET model enhances accuracy in detecting early Seronegative Hashimoto's thyroiditis over senior ultrasound physicians. Ethics No. [2023] H013 Trial registration Chinese Clinical Trial Registry;CTR2400092179; 12 November 2024.https://doi.org/10.1186/s12865-025-00708-5RadiomicsHashimoto’s thyroiditis(HT)Seronegative Hashimoto thyroiditisMachine learningUltrasound imagingHypothyroidism |
| spellingShingle | Wenjun Wu Shengsheng Yao Daming Liu Yuan Luo Yihan Sun Ting Ruan Mengyou Liu Li Shi Chang Liu Mingming Xiao Qi Zhang Zhengshuai Liu Xingai Ju Jiahao Wang Xiang Fei Li Lu Yang Gao Ying Zhang Liying Gong Xuanyu Chen Wanli Zheng Xiali Niu Xiao Yang Huimei Cao Shijie Chang Jianchun Cui Zuoxin Ma Prediction of Seronegative Hashimoto's thyroiditis using machine learning models based on ultrasound radiomics: a multicenter study BMC Immunology Radiomics Hashimoto’s thyroiditis(HT) Seronegative Hashimoto thyroiditis Machine learning Ultrasound imaging Hypothyroidism |
| title | Prediction of Seronegative Hashimoto's thyroiditis using machine learning models based on ultrasound radiomics: a multicenter study |
| title_full | Prediction of Seronegative Hashimoto's thyroiditis using machine learning models based on ultrasound radiomics: a multicenter study |
| title_fullStr | Prediction of Seronegative Hashimoto's thyroiditis using machine learning models based on ultrasound radiomics: a multicenter study |
| title_full_unstemmed | Prediction of Seronegative Hashimoto's thyroiditis using machine learning models based on ultrasound radiomics: a multicenter study |
| title_short | Prediction of Seronegative Hashimoto's thyroiditis using machine learning models based on ultrasound radiomics: a multicenter study |
| title_sort | prediction of seronegative hashimoto s thyroiditis using machine learning models based on ultrasound radiomics a multicenter study |
| topic | Radiomics Hashimoto’s thyroiditis(HT) Seronegative Hashimoto thyroiditis Machine learning Ultrasound imaging Hypothyroidism |
| url | https://doi.org/10.1186/s12865-025-00708-5 |
| work_keys_str_mv | AT wenjunwu predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT shengshengyao predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT damingliu predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT yuanluo predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT yihansun predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT tingruan predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT mengyouliu predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT lishi predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT changliu predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT mingmingxiao predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT qizhang predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT zhengshuailiu predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT xingaiju predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT jiahaowang predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT xiangfei predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT lilu predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT yanggao predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT yingzhang predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT liyinggong predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT xuanyuchen predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT wanlizheng predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT xialiniu predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT xiaoyang predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT huimeicao predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT shijiechang predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT jianchuncui predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy AT zuoxinma predictionofseronegativehashimotosthyroiditisusingmachinelearningmodelsbasedonultrasoundradiomicsamulticenterstudy |