Search

Objective: Large volume radiological text data have been accumulated since the incorporation of electronic health record (EHR) systems in clinical practice. We aimed to determine whether deep natural language processing algorithms could aid radiologists in improving thyroid cancer diagnosis. Methods: Sonographic EHR data were obtained from the EHR database. Pathological reports were used as the gold standard for diagnosing thyroid cancer. We developed thyroid cancer diagnosis based on natural language processing (THCaDxNLP) to interpret unstructured sonographic text reports for thyroid cancer diagnosis. We used the area under the receiver operating characteristic curve (AUROC) as the primary metric to measure the performance of the THCaDxNLP. We compared the performance of thyroid ultrasound radiologists aided with THCaDxNLP vs. those without THCaDxNLP using 5 independent test sets. Results: We obtained a total number of 788,129 sonographic radiological reports. The number of thyroid sonographic data points was 132,277, 18,400 of which were thyroid cancer patients. Among the 5 test sets, the numbers of patients per set were 439, 186, 82, 343, and 171. THCaDxNLP achieved high performance in identifying thyroid cancer patients (the AUROC ranged from 0.857–0.932). Thyroid ultrasound radiologists aided with THCaDxNLP achieved significantly higher performances than those without THCaDxNLP in terms of accuracy (93.8% vs. 87.2%; one-sided t-test, adjusted P = 0.003), precision (92.5% vs. 86.0%; P = 0.018), and F1 metric (94.2% vs. 86.4%; P = 0.007). Conclusions: THCaDxNLP achieved a high AUROC for the identification of thyroid cancer, and improved the accuracy, sensitivity, and precision of thyroid ultrasound radiologists. This warrants further investigation of THCaDxNLP in prospective clinical trials.

Hashimoto’s thyroiditis (HT) is the main cause of hypothyroidism. We develop a deep learning model called HTNet for diagnosis of HT by training on 106,513 thyroid ultrasound images from 17,934 patients and test its performance on 5051 patients from 2 datasets of static images and 1 dataset of video data. HTNet achieves an area under the receiver operating curve (AUC) of 0.905 (95% CI: 0.894 to 0.915), 0.888 (0.836–0.939) and 0.895 (0.862–0.927). HTNet exceeds radiologists’ performance on accuracy (83.2% versus 79.8%; binomial test, p p p = 0.004) and static-image (AUC, 0.914 versus 0.901; p = 0.08) testing sets, respectively. HTNet may be helpful as a tool for the management of HT.

Background We compared the ultrasound features, superb microvascular imaging (SMI) and micro vessel density (MVD) of pleomorphic adenoma (PA), Warthin’s tumor (WT) and basal cell adenoma (BCA) to explore the clinic value of SMI in differential diagnosis of benign tumors of parotid gland. Methods The vascular distributions and grade by color doppler flow imaging (CDFI) and SMI, as well as vascular index (VI) of 249 parotid gland masses from 217 patients were analyzed. Results The internal echogenicity of BCA are more homogeneous in comparing with WT and PA(P P P P Conclusion SMI can provide low-velocity blood flow information, which is helpful for the differential diagnosis of common benign tumors of parotid gland, and is expected to be more widely used.

Purpose To investigate the capacity of Superb Microvascular Imaging (SMI) to detect microvascular details and to explore the different SMI features in various focal liver lesions (FLLs) and the correlation between SMI and microvessel density (MVD). Method: Eighty-three liver lesions were enrolled in our study, including 35 hepatocellular carcinomas (HCCs) and 48 non-HCCs. All patients underwent color Doppler flow imaging (CDFI) and SMI examination and were categorized into subgroups according to Adler semiquantitative grading (grade 0–3) or the microvascular morphologic patterns (pattern a-f). The correlation between SMI blood flow signal percentage and MVD was assessed. Results Compared with CDFI, SMI detected more high-level blood flow signals (grade 2–3) and more hypervascular supply patterns (pattern e-f) in HCCs (p p r = 0.675, p Conclusion SMI was superior to CDFI in detecting microvascular blood flow signals. More hypervascular supply patterns were depicted in HCC than in non-HCC, suggesting a promising diagnostic value for SMI in the differentiation between HCC and non-HCC. Meanwhile, we were the first to demonstrate that SMI blood flow signal percentage (VI) was correlated with MVD in malignant lesions.

BACKGROUND The aim of this study was to evaluate the clinical diagnostic value of superb microvascular imaging (SMI) in assessing vascular distribution, vascularity, and vessel morphology of parotid tumors (PTs). MATERIAL AND METHODS PT patients confirmed by postoperative histopathological detection and who underwent color Doppler flow imaging (CDFI), microvascular imaging (MVI), and SMI examination were recruited. PTs were classified into 3 groups: pleomorphic adenoma (PA), Warthin tumor (WT), and malignant PT (MT). The tumor vascular distribution, vascularity, and vessel morphology recorded by CDFI, MVI, and SMI were compared among PA, WT, and MT group. PT diagnosis was performed using histopathological detection. Fisher's exact test was used to compare the diagnostic sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV), and accuracy between SMI and MVI examination in PTs. RESULTS We enrolled 198 PTs consisting of 114 PAs, 56 WTs, and 28 MTs into our study. CDFI examination found no significant differences in vascular distribution and vascularity among the PA, WT, and WT groups. SMI examination found significant differences in vascular distribution and vascularity among the 3 groups. MVI found significant differences in vessel morphology, including uneven distribution of blood flow, arborization, and irregular blood flow among the PA, WT, and MT groups. SMI found significant differences in arborization and irregular blood flow, but none of the differences in uneven distribution of blood flow among the 3 groups were significant. The diagnostic sensitivity, specificity, and accuracy of SMI and MVI in PTs showed no significant differences. CONCLUSIONS SMI more accurately evaluated the vascular distribution and vascularity of PTs than CDFI. SMI might be a potential non-invasive diagnostic method for PTs in clinical practice.

Ultrasound (US) is the most important imaging in the preoperative diagnosis of medullary thyroid carcinoma (MTC). MTC are easy to be misdiagnosed due to lacking typical malignant US features. This study investigated US features, clinical characteristics, prognosis, and detection methods, aimed to explore the association between US features and biological behavior, and improve early diagnosis of MTC.A total of 189 MTC patients were enrolled in the study. Based on US features, 29 MTC were categorized as "indeterminate" (i-MTC) and 160 MTC were categorized as "malignant" (m-MTC) according to Thyroid Imaging, Reporting and Data System published by America College of Radiology (ACR TI-RADS). We compared US features, clinical characteristics and prognosis between both groups. We analyzed cytological categories of fine needle aspiration (FNA) within each i-MTC and m-MTC group according to the 2017 Bethesda System for Reporting Thyroid Cytopathology (TBSRTC). We assessed the positive rate of FNA, frozen pathological examination, and preoperative serum calcitonin (Ctn) level in i-MTC and m-MTC groups.Preoperative US features were significantly different in shape, margin, composition, echogenicity, and calcifications between i-MTC and m-MTC (p 0.05). I-MTC showed a hypoechoic solid or solid-cystic nodule lacking malignant US features. While m-MTC was presented as a solid nodule with obviously malignant US features. There were significant differences in lymph node dissection, extent of tumor, lymph node metastasis, and TNM stage and prognosis between i-MTC and m-MTC (p 0.05). Compared to m-MTC, i-MTC underwent central neck dissection more frequently rather than lateral neck dissection at the time of the initial operation; i-MTC had less extrathyroidal invasion and lymph node metastasis, earlier stage, higher rate of biochemical cure, and lower rate of structural persistence/recurrence (p 0.05). The 2017 TBSRTC of i-MTC and m-MTC was significantly different (p 0.05). Preoperative serum Ctn level had a higher diagnostic sensitivity for both i-MTC and m-MTC when comparing to FNA and frozen pathological examination (p 0.05).US features were associated with biological characteristics and prognosis of MTC. I-MTC lack malignant US features, preformed less aggressiveness, and better prognosis. TBSRTC according to FNA combined with serum Ctn were helpful for the detection of i-MTC.

Kidney tumor is one of the diseases threatening human health. Ultrasound is widely applied in kidney tumor diagnosis due to its high popularization, low price and no radiation. Accurate segmentation of kidney tumor is the basis of precise treatment. Kidney tumors often grow in the middle of cortex, so that segmentation is easy disturbed by nearby organs. Besides, ultrasound images own low contrast and large speckle, leading to difficult segmentation. This paper proposed a novel kidney tumor segmentation method in ultrasound images using adaptive sub-regional evolution level set models (ASLSM). Regions of interest are firstly divided into subareas. Secondly, object function is designed by integrating inside and outside energy and gradient, in which the ratio of these two parts are adjusted adaptively. Thirdly, ASLSM adapts convolution radius and curvature according to centroid principle and similarity inside and outside zero level set. Hausdorff distance (HD) of (8.75 ± 4.21) mm, mean absolute distance (MAD) of (3.26 ± 1.69) mm, dice-coefficient (DICE) of 0.93 ± 0.03 were obtained in the experiment. Compared with traditional ultrasound segmentation method, ASLSM is more accurate in kidney tumor segmentation. ASLSM may offer convenience for doctor to locate and diagnose kidney tumor in the future.肾脏肿瘤已经成为威胁人类健康的重要疾病之一。超声检查具有普及率高、价格低廉、无辐射等诸多优点，已广泛应用于肾脏肿瘤的诊断中。超声图像中肾脏肿瘤的准确分割是制定治疗方案的基础。肾脏肿瘤往往生长在肾皮质中间，分割容易受到周围脏器干扰，而且超声图像对比度低、斑点噪声严重，使得肿瘤分割困难。本文根据肾脏超声图像的特点，提出基于自适应分区演化水平集（ASLSM）的肿瘤分割算法。首先，将感兴趣区域图像分区；然后，融合内外能量项和梯度设计目标函数，并自适应调整二者比例；最后，根据质心原理和零水平集内外相似度自适应卷积半径及曲率，进行曲线演化。将本算法用于肾脏超声图像，实验结果中豪斯多夫距离（HD）为（8.75 ± 4.21）mm，平均绝对距离（MAD）为（3.26 ± 1.69）mm，戴斯系数（DICE）为 0.93 ± 0.03。与传统的方法进行比较，实验结果证明本算法可以获得更加准确的肿瘤分割结果，今后本算法或可为辅助医生定位和诊断肾脏肿瘤提供便利。.

Objective: Bosniak classification system provides a fine imaging evaluation for the malignant cystic renal masses. Superb microvascular imaging (SMI) is a new ultrasonic technique which apply advanced clutter suppression to reflect microflow information. The aim of this study was to evaluate the performance of ultrasound Bosniak classification aided by SMI in diagnosis of cystic renal masses. Methods: By comparing with contrast-enhanced ultrasound (CEUS) and pathology, we evaluated the sensitivity, specificity and accuracy of conventional ultrasound and SMI combination in the diagnosis of renal cystic masses. Results: Color Doppler Flow image (CDFI) and SMI had significant difference in evaluation the blood flow and Bosniak classification for cystic renal mass of IIf, III and IV (p = 0.020). SMI was consistent with CEUS for cystic renal mass Bosniak classification of IIf, III and IV. Although, CEUS had highest sensitivity of 97.4%, SMI had the highest area under curve in differential diagnosis malignance from benign masses [area under curve = 0.869 (0.743 – 0.995), p < 0.001]. Conclusion: SMI can display better in observing blood flow in septa and solid structures of renal cystic lesions compared with CDFI, thus, it can perform more accurate Bosniak classification for renal cystic mass. SMI and CEUS are consistent in Bosniak classification. Bosniak classification aided by SMI maybe an accurate non-invasive ultrasonic examination in distinguishing benign and malignant renal cystic lesions. Advances in knowledge: Conventional ultrasound had limited ability to distinguish malignance in Bosniak classifications IIf and III. SMI can display better in observing blood flow in septa and solid structures of renal cystic lesions compared with CDFI.

Gastric cancer remains a disease with a high mortality rate despite of multiple therapeutic strategies. So far, it is very important to develop new treatment approaches to improve current therapeutic efficacy in gastric cancer. Apurinic/apyrimidinic endonuclease (APE1) involves in DNA base excision repair (BER) during DNA damage pathway. APE1 was found to be associated with poor overall survival with gastric cancer patients. In the in vitro experiment, we tested APE1 inhibitor-AT101 could potently inhibit gastric cancer cell growth and further induce cancer cell apoptosis and autophagy through p53-dependent pathway. Downregulation of APE1 by AT101 has ability to suppress gastric cancer cell migration and renewal through inhibition of CD133, Nanog and LC3expression. Based on findings that Her-2 positive expression cases has poor prognosis from our dataset and TCGA database, we investigated the role of AT101 in synergetic efficacy with 5-FU treatment in Her-2 overexpression gastric cancer in vivo, indicating that AT101 is able to enhance 5-FU in the shrinkage of xenograft mice tumor and induction of cell apoptosis. In summary, the data obtained from our study showed APE1 is guided as a potential therapeutic target for gastric cancer. AT101 could be regarded as a potent inhibitor to promote chemotherapeutic sensitivity in patients with gastric cancer.

The DNA base excision repair gene APE1 involves in DNA damage repair pathway and overexpression in a variety of human cancers. Analyses of patients with non-small cell lung cancer (NSCLC) suggested that multiple factors associated with prognosis of NSCLC patients. Further investigation showed that APE1 expression was able to predict the progression-free survival and overall survival in patients with NSCLC and correlated with lymph node metastasis. Intriguingly, as a stratification of APE1-141 SNPs in APE1 positive expression, we also found APE1-141 GT/GG was identified as a marker for prediction of poor survival in NSCLC patients. In the in vitro experiments, the results showed that when APE1 expression was inhibited by siRNA or AT101 (an APE1 inhibitor), the migration and invasion of NSCLC cells were suppressed. Furthermore, Epithelial-Mesenchymal Transition (EMT) markers was tested to provide evidence that APE1 promoted NSCLC EMT through interaction with SirT1. Using NSCLC xenograft models, we confirmed that AT101 shrank tumor volumes and inhibited lymph node metastasis. In conclusion, APE1 could be a potential target for patients with NSCLC metastasis and AT101 is a potent inhibitor in further treatment of NSCLC patients.

This study is to evaluate the value of superb microvascular imaging (SMI) in assessing renal tumour vascularity and in differentiating benign from malignant renal tumours.Patients with pathologically confirmed renal mass who underwent colour Doppler flow imaging (CDFI) and SMI examination between October 2016 and March 2017 were retrospectively analysed. Patients were classified into benign renal mass group and malignant tumour group according to their pathological results. The vascularity, vascular distribution and vessel morphology obtained from the two scanning methods were analysed and compared between the two groups.53 patients with renal tumour were enrolled into the study, including 11 cases of benign renal tumours and 42 cases of malignant ones. SMI showed significant difference in vascularity between the two groups; CDFI showed no statistical difference. Compared with CDFI, SMI showed significantly higher grade in malignant tumour (p0.001), and no significant difference (p = 0.421) in benign mass. According to CDFI, there is no statistical difference in vascular distribution and vessel morphology (p = 0.432, p = 0.122) between the two groups, while SMI method showed significant difference (p0.05) between the two groups. SMI was significantly more sensitive and accurate than CDFI.Comparing with CDFI, SMI could effectively detect vascularity, vascular characteristics and distribution and annular blood flow in renal tumour. SMI appears more sensitive and precise in distinguishing benign renal mass from malignant tumour. Advances in knowledge: SMI seems potentially valuable in evaluating renal tumour vascularity and in differentiating benign from malignant renal tumours.

Background Incidence rate of thyroid cancer is steadily increasing due to overdiagnosis and overtreatment. Thyroid ultrasound is commonly used to diagnose thyroid cancer. The aim of this study is to examine the accuracy of using deep convolutional neural network (DCNN) models to improve diagnosis of thyroid cancer by analyzing sonographic imaging data from clinical thyroid ultrasound. Methods A total of 131,731 sonographic images from 17,627 thyroid cancer patients and 180,668 sonographic images from 25,325 controls used as training set were obtained from Tianjin Cancer Hospital. Images from anatomical sites that did not have cancer according to location sign on the image were not included. All thyroid cancer patients and 13·2% of controls (51,255 images) were confirmed by pathological reports. DCNN is a specific type of neural network optimized for image recognition. We trained two DCNN models on the training set and subsequently evaluated the performance on one independent internal (Tianjin, 1,118 individuals) and two external (Jilin,154 individuals; Weihai, 1,420 individuals) validation sets. Individuals in the validation sets all have pathological examinations. We compared the specificity/sensitivity of DCNN models with the performance of six thyroid ultrasound radiologists on these three validation sets. Findings DCNN model achieved high performance in identifying thyroid cancer patients versus six experience radiologists: for Tianjin validation set, sensitivity was 92·2% versus 96·9% (95% CI 89·7% - 94·3% vs. 93·9% - 98·6%; p = 0·003), and specificity was 85·6% versus 59·4% (95% CI 82·4% - 88·4% vs. 53% - 65·6%; p < 0·0001); for Jilin validation set, sensitivity was 84·3% versus 92·9% (95% CI 73·6% - 91·9% vs. 84·1% - 97·6%; p = 0·05), and specificity was 86·9% versus 57·1% (95% CI 77·8% - 93·3% vs. 45·9% - 67·9%; p < 0·0001); for Weihai validation set, sensitivity was 84·5% versus 89% (95% CI 81·2% - 87·4% vs. 81·9% - 94%; p = 0·2), and specificity was 87·5% versus 68·6% (95% CI 85·1% - 89·6% vs. 60·7% - 75·8%; p < 0·0001). Interpretation DCNN models exhibited high accuracy, sensitivity, and specificity in identifying thyroid cancer patients at levels comparable to or higher than six experienced radiologists. Conferred by the high specificity of DCNN models, the rate of overdiagnosis and overtreatment of patients with thyroid cancer is expected to decrease. This supports future application of the deep learning models to clinical practice for thyroid cancer diagnosis. However, further validation of these DCNN models in prospective clinical trials is warranted. Funding The Program for Changjiang Scholars and Innovative Research Team in University in China (IRT_14R40), National Natural Science Foundation of China (31801117). Citation Format: Xiangchun Li, Sheng Zhang, Qiang Zhang, Xi Wei, Yi Pan, Jing Zhao, Xiaojie Xin, Xiaoqing Wang, Fan Yang, Jianxin Li, Meng Yang, Qinghua Wang, Xiangqian Zheng, Yanhui Zhao, Lun Zhang, Xudong Wang, Zhimin Zheng, Christopher T. Whitlow, Metin N. Gurcan, Boris Pasche, Ming Gao, Wei Zhang, Kexin Chen. Diagnosis of thyroid cancer using deep convolutional neural network models applied to sonographic images from clinical ultrasound exams [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 1394.

Ultrasound-guided biopsy technique with the large-core needle has widely been applied in the diagnosis of adult abdominopelvic cavity, thyroid, and neck tumors. There are few reports on ultrasound-guided biopsy using large-core needle in pediatric abdominopelvic cavity tumors. This study was to evaluate the ultrasound features and the diagnostic value of ultrasound-guided core needle biopsy for pediatric neuroblastic tumors. The pediatric patients with neuroblastic tumor that underwent ultrasound examination and ultrasound-guided core needle biopsy from January 2009 to November 2015 were reviewed. A minimum of two cores in each case was obtained. The biopsy results were confirmed by subsequent surgical histopathology. The ultrasound features and the diagnostic accuracy of ultrasound-guided core needle biopsy were evaluated. Eighty-three patients were enrolled into the study. Conventional ultrasound examination showed irregular hypoechoic or mixed echo masses and calcification and liquefied necrosis. The diagnostic accuracy of ultrasound-guided core needle biopsy was 96.4% (80/83). Three cases were misdiagnosed because of inadequate tissue sample. No serious complication, infection, or needle track seeding occurred. Ultrasound-guided core needle biopsy seems to be an accurate, minimally invasive, and safe diagnostic method of pediatric neuroblastic tumor.

To seek for the ultrasound findings of thyroid nodules related to malignancy and benign, and to assess the role of TI-RADS in the ultrosound evaluation of thyroid nodules.We reviewed and analyzed the ultrasound characteristics of 1838 thyroid nodules confirmed by biopsy or surgical pathology, classified the thyroid nodules by TI-RADS grading criteria, and analyzed the malignancy rate and accuracy rate of different TI-RADS grading.Among the 1 838 thyroid nodules, 1 160 cases were carcinomas confirmed by surgery, while benign nodules in 212 cases. The sensitivity, specificity, PPV, NPV, and accuracy rates were 99.7% (1 156/1 160), 41.0%(278/678), 74.3% (1 156/1 556), 98.6% (278/282), and 78.2% (1 434/1 838), respectively. There were significant differences between the malignant and benign thyroid nodules in echo, margin, shape, calcification, and blood flow (P0.001). Hypoechogenicity, marked hypoechogenicity, ill-defined margin, microcalcification, a taller-than-wide shape, and nodule internal rich flow (type III) were significantly associated with malignancy, while hyper/isoechogenicity, smooth margin, macro/no-calcifications, a wider-than-tall shape, and nodules internal poor flow (type I or type II) were significantly associated with benign nodules. There were significant differences between the malignant rates and accuracy rates obtained by different TI-RADS classifications (P0.01).Understanding the ultrasound characteristics of benign and malignant thyroid nodules and applying TI-RADS grading criteria to correctly classify the thyroid nodules are crucial for the clinical treatment and prognosis.

To explore the ultrasound (US) findings of partially cystic thyroid nodules (PCTNs) and to analyze their relationship with differential diagnosis of benign and malignant lesions.265 cases of PCTNs confirmed by needle biopsy or surgical pathology were included in this study. Their ultrasound characteristics were reviewed and their significance in differential diagnosis was analyzed.In the 265 PTCNs, 53 cases were malignant and 212 cases were benign lesions. According to the comparison of ultrasound and pathology results, there were 51 true-positive cases, 208 true negative cases, four false-positive cases and two false negative cases;the sensitivity was 96.2% (51/53), specificity was 98.1% (208/212), positive predictive value (PPV) was 92.7% (51/55), negative predictive values (NPV) was 99.0% (208/210), and accuracy rate was 97.7% (259/265). In terms of the PTCNs, internal structure, shape and margin were significantly associated with malignant or benign nature (P0.001); the aspect ratio ≥ 1, spiculated or micro-lobulated margin were significantly associated with malignancy; while the smooth margin, spongiform structure were significantly associated with benign nature. In terms of the internal solid portion of the nodules, the configuration, free margin, echogenecity, and calcification were significantly associated with malignant or benign nature (P0.001). The eccentric configuration with an acute angle, non-smooth free margin, hypoechogenecity, marked hypoechogenecity and micro-calcification were significantly associated with malignancy; while concentric configuration, smooth free margin, hyper/isoechogenicity, free margin of the solid component, and macro-calcifications were significantly associated with benign nature (P0.01). The results of logistic regression analysis showed that the echogenicity, free margin, configuration and calcification of the solid component were important predictive factors of malignant lesions (P0.05), that the hypoechogenecity, marked hypoechogenecity, eccentric configuration with an acute angle, non-smooth free margin and micro-calcification of the internal solid portion of the nodules were predictors for malignant PTCNs (P0.01).Understanding the characteristics of US findings of partially cystic thyroid nodules is of great importance to make an accurate diagnosis of malignant nodules.

To assess the diagnostic value of elastosonography for thyroid microcarcinoma (TMC), particularly with regard to elasticity score (ES) and strain ratio (SR).Conventional ultrasound and elastosonography were performed for 487 thyroid micronodules before surgery. We set the histology as the reference standard. The ES and SR values, as well as their diagnostic threshold and efficiency, were compared and analyzed by the receiver-operating characteristic (ROC) curve. Additional comparisons between TMC patients with and without extracapsular extension were also performed.Statistically significant differences (P0.05) in both ES and SR values were detected among the TMC and benign groups. The area under the ROC curve of SR was significantly greater than that of ES (0.956 and 0.844, respectively; P0.05). Using ES⩾3 and SR⩾3.65 as diagnostic threshold values, the diagnostic sensitivity, specificity, and accuracy of ES for differentiating benign and malignant nodules were 79.9%, 72.3%, and 80.5%, respectively, whereas those of SR were 86.6%, 85.3%, and 89.4%, respectively. The maximum diameter, microcalcification status, aspect ratio, bilateral cervical lymph node metastasis, and SR values of nodules with extracapsular extension (A1 subgroup) were greater than those of nodules without extracapsular extension (A2 subgroup).Elasticity imaging technology not only can help differentiate between benign and malignant thyroid micronodules but also allow SR values to provide accurate and objective information on tissue hardness and to predict TMC extracapsular extension or even bilateral cervical lymph node metastasis.