Tongtong Liu1,2, Xifeng Ge3, Jinhua Yu4,5, Yi Guo1,2, Yuanyuan Wang1,2, Wenping Wang6, Ligang Cui7. 1. Department of Electronic Engineering, Fudan University, Shanghai, 200433, China. 2. Key Laboratory of Medical Imaging, Computing and Computer-Assisted Intervention, Shanghai, 200433, China. 3. Department of Ultrasound, Peking University Third Hospital, Beijing, 100191, China. 4. Department of Electronic Engineering, Fudan University, Shanghai, 200433, China. jhyu@fudan.edu.cn. 5. Key Laboratory of Medical Imaging, Computing and Computer-Assisted Intervention, Shanghai, 200433, China. jhyu@fudan.edu.cn. 6. Department of Ultrasound, Zhongshan Hospital, Fudan University, Shanghai, 20032, China. 7. Department of Ultrasound, Peking University Third Hospital, Beijing, 100191, China. cuiligang_bysy@126.com.
Abstract
PURPOSE: B-mode ultrasound (B-US) and strain elastography ultrasound (SE-US) images have a potential to distinguish thyroid tumor with different lymph node (LN) status. The purpose of our study is to investigate whether the application of multi-modality images including B-US and SE-US can improve the discriminability of thyroid tumor with LN metastasis based on a radiomics approach. METHODS: Ultrasound (US) images including B-US and SE-US images of 75 papillary thyroid carcinoma (PTC) cases were retrospectively collected. A radiomics approach was developed in this study to estimate LNs status of PTC patients. The approach included image segmentation, quantitative feature extraction, feature selection and classification. Three feature sets were extracted from B-US, SE-US, and multi-modality containing B-US and SE-US. They were used to evaluate the contribution of different modalities. A total of 684 radiomics features have been extracted in our study. We used sparse representation coefficient-based feature selection method with 10-bootstrap to reduce the dimension of feature sets. Support vector machine with leave-one-out cross-validation was used to build the model for estimating LN status. RESULTS: Using features extracted from both B-US and SE-US, the radiomics-based model produced an area under the receiver operating characteristic curve (AUC) [Formula: see text] 0.90, accuracy (ACC) [Formula: see text] 0.85, sensitivity (SENS) [Formula: see text] 0.77 and specificity (SPEC) [Formula: see text] 0.88, which was better than using features extracted from B-US or SE-US separately. CONCLUSIONS: Multi-modality images provided more information in radiomics study. Combining use of B-US and SE-US could improve the LN metastasis estimation accuracy for PTC patients.
PURPOSE: B-mode ultrasound (B-US) and strain elastography ultrasound (SE-US) images have a potential to distinguish thyroid tumor with different lymph node (LN) status. The purpose of our study is to investigate whether the application of multi-modality images including B-US and SE-US can improve the discriminability of thyroid tumor with LN metastasis based on a radiomics approach. METHODS: Ultrasound (US) images including B-US and SE-US images of 75 papillary thyroid carcinoma (PTC) cases were retrospectively collected. A radiomics approach was developed in this study to estimate LNs status of PTC patients. The approach included image segmentation, quantitative feature extraction, feature selection and classification. Three feature sets were extracted from B-US, SE-US, and multi-modality containing B-US and SE-US. They were used to evaluate the contribution of different modalities. A total of 684 radiomics features have been extracted in our study. We used sparse representation coefficient-based feature selection method with 10-bootstrap to reduce the dimension of feature sets. Support vector machine with leave-one-out cross-validation was used to build the model for estimating LN status. RESULTS: Using features extracted from both B-US and SE-US, the radiomics-based model produced an area under the receiver operating characteristic curve (AUC) [Formula: see text] 0.90, accuracy (ACC) [Formula: see text] 0.85, sensitivity (SENS) [Formula: see text] 0.77 and specificity (SPEC) [Formula: see text] 0.88, which was better than using features extracted from B-US or SE-US separately. CONCLUSIONS: Multi-modality images provided more information in radiomics study. Combining use of B-US and SE-US could improve the LN metastasis estimation accuracy for PTC patients.